Rob J Hyndman. Visualizing and forecasting. big time series data. Victoria: scaled
|
|
- Janis Johnson
- 5 years ago
- Views:
Transcription
1 Rob J Hyndman Visualizing and forecasting big time series data Victoria: scaled
2 Outline 1 Examples of biggish time series 2 Time series visualisation 3 BLUF: Best Linear Unbiased Forecasts 4 Application: Australian tourism 5 Fast computation tricks 6 hts package for R 7 References Visualising and forecasting big time series data Examples of biggish time series 2
3 1. Australian tourism demand Visualising and forecasting big time series data Examples of biggish time series 3
4 1. Australian tourism demand Quarterly data on visitor night from 1998:Q1 2013:Q4 From: National Visitor Survey, based on annual interviews of 120,000 Australians aged 15+, collected by Tourism Research Australia. Split by 7 states, 27 zones and 76 regions (a geographical hierarchy) Also split by purpose of travel Holiday Visiting friends and relatives (VFR) Business Other 304 bottom-level series Visualising and forecasting big time series data Examples of biggish time series 3
5 2. Labour market participation Australia and New Zealand Standard Classification of Occupations 8 major groups 43 sub-major groups 97 minor groups 359 unit groups * 1023 occupations Example: statistician 2 Professionals 22 Business, Human Resource and Marketing Professionals 224 Information and Organisation Professionals 2241 Actuaries, Mathematicians and Statisticians Statistician Visualising and forecasting big time series data Examples of biggish time series 4
6 2. Labour market participation Australia and New Zealand Standard Classification of Occupations 8 major groups 43 sub-major groups 97 minor groups 359 unit groups * 1023 occupations Example: statistician 2 Professionals 22 Business, Human Resource and Marketing Professionals 224 Information and Organisation Professionals 2241 Actuaries, Mathematicians and Statisticians Statistician Visualising and forecasting big time series data Examples of biggish time series 4
7 3. Spectacle sales Monthly UK sales data from Provided by a large spectacle manufacturer Split by brand (26), gender (3), price range (6), materials (4), and stores (600) About 1 million bottom-level series Visualising and forecasting big time series data Examples of biggish time series 5
8 3. Spectacle sales Monthly UK sales data from Provided by a large spectacle manufacturer Split by brand (26), gender (3), price range (6), materials (4), and stores (600) About 1 million bottom-level series Visualising and forecasting big time series data Examples of biggish time series 5
9 3. Spectacle sales Monthly UK sales data from Provided by a large spectacle manufacturer Split by brand (26), gender (3), price range (6), materials (4), and stores (600) About 1 million bottom-level series Visualising and forecasting big time series data Examples of biggish time series 5
10 3. Spectacle sales Monthly UK sales data from Provided by a large spectacle manufacturer Split by brand (26), gender (3), price range (6), materials (4), and stores (600) About 1 million bottom-level series Visualising and forecasting big time series data Examples of biggish time series 5
11 Outline 1 Examples of biggish time series 2 Time series visualisation 3 BLUF: Best Linear Unbiased Forecasts 4 Application: Australian tourism 5 Fast computation tricks 6 hts package for R 7 References Visualising and forecasting big time series data Time series visualisation 6
12 Kite diagrams Line graph profile 0 0 Duplicate & flip around the horizontal axis 0 Fill the colour Visualising and forecasting big time series data Time series visualisation 7
13 Kite diagrams: Victorian tourism Victoria BAA BAB BAC BBA BCA BCB BCC BDA BDB BDC BDD BDE BDF BEA BEB BEC BED BEE BEF BEG Other Business VFR Holiday Visualising and forecasting big time series data Time series visualisation 8
14 Kite diagrams: Victorian tourism Visualising and forecasting big time series data Time series visualisation 8
15 Kite diagrams: Victorian tourism Visualising and forecasting big time series data Time series visualisation 8
16 Kite diagrams: Victorian tourism Victoria: scaled BAA BAB BAC BBA BCA BCB BCC BDA BDB BDC BDD BDE BDF BEA BEB BEC BED BEE BEF BEG Other Business VFR Holiday Visualising and forecasting big time series data Time series visualisation 8
17 An STL decomposition STL decomposition of tourism demand for holidays in Peninsula remainder trend seasonal data Visualising and forecasting big time series data Time series visualisation 9
18 Seasonal stacked bar chart Place positive values above the origin while negative values below the origin Map the bar length to the magnitude Encode quarters by colours Visualising and forecasting big time series data Time series visualisation 10
19 Seasonal stacked bar chart Place positive values above the origin while negative values below the origin Map the bar length to the magnitude Encode quarters by colours 1.0 Seasonal Component Holiday Qtr Q1 Q2 Q3 Q4 BAA BABBACBBABCABCBBCCBDABDBBDCBDDBDEBDF BEA BEBBECBEDBEE BEFBEG Regions Visualising and forecasting big time series data Time series visualisation 10
20 Seasonal stacked bar chart: VIC Visualising and forecasting big time series data Time series visualisation 11
21 Corrgram of remainder Compute the correlations among the remainder components Render both the sign and magnitude using a colour mapping of two hues Order variables according to the first principal component of the correlations. Visualising and forecasting big time series data Time series visualisation 12
22 Corrgram of remainder: VIC Visualising and forecasting big time series data Time series visualisation BEEHol BEFOth BEEOth BDEOth BEBOth BEABus BEFBus BDCOth BACHol BEBBus BEAVis BBAHol BDEHol BABOth BAAVis BAAHol BDCHol BBABus BCBHol BEGBus BDDVis BABVis BDAVis BEAOth BDFHol BEEBus BAAOth BACOth BDAOth BDEBus BCBOth BACBus BEBVis BACVis BCAOth BEFVis BCBVis BEDHol BEGOth BDBHol BABBus BEBHol BDFBus BECHol BCAHol BDBOth BEAHol BDCBus BECVis BDBVis BCCHol BBAVis BABHol BBAOth BCCOth BCBBus BCCVis BEGVis BDDHol BECOth BDCVis BAABus BCCBus BECBus BCAVis BDFVis BEGHol BDDOth BEDOth BEDVis BDDBus BDEVis BEFHol BEEVis BDBBus BDABus BDAHol BCABus BDFOth BEDBus BEEHol BEFOth BEEOth BDEOth BEBOth BEABus BEFBus BDCOth BACHol BEBBus BEAVis BBAHol BDEHol BABOth BAAVis BAAHol BDCHol BBABus BCBHol BEGBus BDDVis BABVis BDAVis BEAOth BDFHol BEEBus BAAOth BACOth BDAOth BDEBus BCBOth BACBus BEBVis BACVis BCAOth BEFVis BCBVis BEDHol BEGOth BDBHol BABBus BEBHol BDFBus BECHol BCAHol BDBOth BEAHol BDCBus BECVis BDBVis BCCHol BBAVis BABHol BBAOth BCCOth BCBBus BCCVis BEGVis BDDHol BECOth BDCVis BAABus BCCBus BECBus BCAVis BDFVis BEGHol BDDOth BEDOth BEDVis BDDBus BDEVis BEFHol BEEVis BDBBus BDABus BDAHol BCABus BDFOth BEDBus
23 Corrgram of remainder: TAS FCAHol FBBHol FBAHol FAAHol FCBHol FCAVis FBBVis FAAVis FCBBus FAAOth FCAOth FBBOth FBABus FBAOth FCBVis FCABus FBAVis FCBOth FBBBus FCAHol FBBHol FBAHol FAAHol FCBHol FCAVis FBBVis FAAVis FCBBus FAAOth FCAOth FBBOth FBABus FBAOth FCBVis FCABus FBAVis FCBOth FBBBus FAABus FAABus 1 Visualising and forecasting big time series data Time series visualisation 14
24 Feature analysis Summarize each time series with a feature vector: strength of trend lumpiness (variance of annual variances of remainder) strength of seasonality size of seasonal peak size of seasonal trough ACF1 linearity of trend curvature of trend spectral entropy Do PCA on feature matrix Visualising and forecasting big time series data Time series visualisation 15
25 Feature analysis standardized PC2 (18.6% explained var.) season trough peak fo.acf linearity curvature lump spike entropy standardized PC1 (43.2% explained var.) Visualising and forecasting big time series data Time series visualisation 16
26 Feature analysis 800 value BEGBus BCCBus CCAOth CACOth Time Visualising and forecasting big time series data Time series visualisation 16
27 Feature analysis 1600 value ADAHol BDCHol BBAHol ACAHol Time Visualising and forecasting big time series data Time series visualisation 16
28 Feature analysis Visualising and forecasting big time series data Time series visualisation 16 NSW NT QLD SA TAS VIC WA PC1 PC2
29 Feature analysis Visualising and forecasting big time series data Time series visualisation 16 Bus Hol Oth Vis PC1 PC2
30 Outline 1 Examples of biggish time series 2 Time series visualisation 3 BLUF: Best Linear Unbiased Forecasts 4 Application: Australian tourism 5 Fast computation tricks 6 hts package for R 7 References Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 17
31 Hierarchical time series A hierarchical time series is a collection of several time series that are linked together in a hierarchical structure. Total A AA AB AC B BA BB BC C CA CB CC Examples Net labour turnover Tourism by state and region Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 18
32 Hierarchical time series A hierarchical time series is a collection of several time series that are linked together in a hierarchical structure. Total A AA AB AC B BA BB BC C CA CB CC Examples Net labour turnover Tourism by state and region Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 18
33 Hierarchical time series A hierarchical time series is a collection of several time series that are linked together in a hierarchical structure. Total A AA AB AC B BA BB BC C CA CB CC Examples Net labour turnover Tourism by state and region Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 18
34 Hierarchical time series Total A B C Y t : observed aggregate of all series at time t. Y X,t : observation on series X at time t. B t : vector of all series at bottom level in time t. Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 19
35 Hierarchical time series Total A B C Y t : observed aggregate of all series at time t. Y X,t : observation on series X at time t. B t : vector of all series at bottom level in time t. Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 19
36 Hierarchical time series Total A B C y t = [Y t, Y A,t, Y B,t, Y C,t ] = Y t : observed aggregate of all series at time t. Y X,t : observation on series X at time t. B t : vector of all series at bottom level in time t Y A,t Y B,t Y C,t Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 19
37 Hierarchical time series Total A B C y t = [Y t, Y A,t, Y B,t, Y C,t ] = Y t : observed aggregate of all series at time t. Y X,t : observation on series X at time t. B t : vector of all series at bottom level in time t Y A,t Y B,t Y C,t }{{} S Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 19
38 Hierarchical time series Total A B C y t = [Y t, Y A,t, Y B,t, Y C,t ] = Y t : observed aggregate of all series at time t. Y X,t : observation on series X at time t. B t : vector of all series at bottom level in time t Y A,t Y B,t Y C,t }{{}}{{} B t S Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 19
39 Hierarchical time series Total A B C y t = [Y t, Y A,t, Y B,t, Y C,t ] = y t = SB t Y t : observed aggregate of all series at time t. Y X,t : observation on series X at time t. B t : vector of all series at bottom level in time t Y A,t Y B,t Y C,t }{{}}{{} B t S Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 19
40 Hierarchical time series Total A B C y t = AX AY AZ Y t Y A,t Y B,t Y C,t Y AX,t Y AY,t Y AZ,t Y BX,t Y BY,t Y BZ,t Y CX,t Y CY,t Y CZ,t = BX BY BZ } {{ } S Y AX,t Y AY,t Y AZ,t Y BX,t Y BY,t Y BZ,t Y CX,t Y CY,t Y CZ,t }{{} B t CX CY CZ Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 20
41 Hierarchical time series Total A B C y t = AX AY AZ Y t Y A,t Y B,t Y C,t Y AX,t Y AY,t Y AZ,t Y BX,t Y BY,t Y BZ,t Y CX,t Y CY,t Y CZ,t = BX BY BZ } {{ } S Y AX,t Y AY,t Y AZ,t Y BX,t Y BY,t Y BZ,t Y CX,t Y CY,t Y CZ,t }{{} B t CX CY CZ Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 20
42 Hierarchical time series Total A B C y t = AX AY AZ Y t Y A,t Y B,t Y C,t Y AX,t Y AY,t Y AZ,t Y BX,t Y BY,t Y BZ,t Y CX,t Y CY,t Y CZ,t = BX BY BZ } {{ } S Y AX,t Y AY,t Y AZ,t Y BX,t Y BY,t Y BZ,t Y CX,t Y CY,t Y CZ,t }{{} B t CX CY CZ y t = SB t Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 20
43 Forecasting notation Let ŷ n (h) be vector of initial h-step forecasts, made at time n, stacked in same order as y t. (They may not add up.) Reconciled forecasts are of the form: ỹ n (h) = SPŷ n (h) for some matrix P. P extracts and combines base forecasts ŷ n (h) to get bottom-level forecasts. S adds them up Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 21
44 Forecasting notation Let ŷ n (h) be vector of initial h-step forecasts, made at time n, stacked in same order as y t. (They may not add up.) Reconciled forecasts are of the form: ỹ n (h) = SPŷ n (h) for some matrix P. P extracts and combines base forecasts ŷ n (h) to get bottom-level forecasts. S adds them up Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 21
45 Forecasting notation Let ŷ n (h) be vector of initial h-step forecasts, made at time n, stacked in same order as y t. (They may not add up.) Reconciled forecasts are of the form: ỹ n (h) = SPŷ n (h) for some matrix P. P extracts and combines base forecasts ŷ n (h) to get bottom-level forecasts. S adds them up Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 21
46 Forecasting notation Let ŷ n (h) be vector of initial h-step forecasts, made at time n, stacked in same order as y t. (They may not add up.) Reconciled forecasts are of the form: ỹ n (h) = SPŷ n (h) for some matrix P. P extracts and combines base forecasts ŷ n (h) to get bottom-level forecasts. S adds them up Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 21
47 Forecasting notation Let ŷ n (h) be vector of initial h-step forecasts, made at time n, stacked in same order as y t. (They may not add up.) Reconciled forecasts are of the form: ỹ n (h) = SPŷ n (h) for some matrix P. P extracts and combines base forecasts ŷ n (h) to get bottom-level forecasts. S adds them up Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 21
48 General properties: bias ỹ n (h) = SPŷ n (h) Assume: base forecasts ŷ n (h) are unbiased: E[ŷ n (h) y 1,..., y n ] = E[y n+h y 1,..., y n ] Let ˆBn (h) be bottom level base forecasts with β n (h) = E[ˆBn (h) y 1,..., y n ]. Then E[ŷ n (h)] = Sβ n (h). We want the revised forecasts to be unbiased: E[ỹ n (h)] = SPSβ n (h) = Sβ n (h). Revised forecasts are unbiased iff SPS = S. Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 22
49 General properties: bias ỹ n (h) = SPŷ n (h) Assume: base forecasts ŷ n (h) are unbiased: E[ŷ n (h) y 1,..., y n ] = E[y n+h y 1,..., y n ] Let ˆBn (h) be bottom level base forecasts with β n (h) = E[ˆBn (h) y 1,..., y n ]. Then E[ŷ n (h)] = Sβ n (h). We want the revised forecasts to be unbiased: E[ỹ n (h)] = SPSβ n (h) = Sβ n (h). Revised forecasts are unbiased iff SPS = S. Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 22
50 General properties: bias ỹ n (h) = SPŷ n (h) Assume: base forecasts ŷ n (h) are unbiased: E[ŷ n (h) y 1,..., y n ] = E[y n+h y 1,..., y n ] Let ˆBn (h) be bottom level base forecasts with β n (h) = E[ˆBn (h) y 1,..., y n ]. Then E[ŷ n (h)] = Sβ n (h). We want the revised forecasts to be unbiased: E[ỹ n (h)] = SPSβ n (h) = Sβ n (h). Revised forecasts are unbiased iff SPS = S. Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 22
51 General properties: bias ỹ n (h) = SPŷ n (h) Assume: base forecasts ŷ n (h) are unbiased: E[ŷ n (h) y 1,..., y n ] = E[y n+h y 1,..., y n ] Let ˆBn (h) be bottom level base forecasts with β n (h) = E[ˆBn (h) y 1,..., y n ]. Then E[ŷ n (h)] = Sβ n (h). We want the revised forecasts to be unbiased: E[ỹ n (h)] = SPSβ n (h) = Sβ n (h). Revised forecasts are unbiased iff SPS = S. Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 22
52 General properties: bias ỹ n (h) = SPŷ n (h) Assume: base forecasts ŷ n (h) are unbiased: E[ŷ n (h) y 1,..., y n ] = E[y n+h y 1,..., y n ] Let ˆBn (h) be bottom level base forecasts with β n (h) = E[ˆBn (h) y 1,..., y n ]. Then E[ŷ n (h)] = Sβ n (h). We want the revised forecasts to be unbiased: E[ỹ n (h)] = SPSβ n (h) = Sβ n (h). Revised forecasts are unbiased iff SPS = S. Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 22
53 General properties: bias ỹ n (h) = SPŷ n (h) Assume: base forecasts ŷ n (h) are unbiased: E[ŷ n (h) y 1,..., y n ] = E[y n+h y 1,..., y n ] Let ˆBn (h) be bottom level base forecasts with β n (h) = E[ˆBn (h) y 1,..., y n ]. Then E[ŷ n (h)] = Sβ n (h). We want the revised forecasts to be unbiased: E[ỹ n (h)] = SPSβ n (h) = Sβ n (h). Revised forecasts are unbiased iff SPS = S. Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 22
54 General properties: bias ỹ n (h) = SPŷ n (h) Assume: base forecasts ŷ n (h) are unbiased: E[ŷ n (h) y 1,..., y n ] = E[y n+h y 1,..., y n ] Let ˆBn (h) be bottom level base forecasts with β n (h) = E[ˆBn (h) y 1,..., y n ]. Then E[ŷ n (h)] = Sβ n (h). We want the revised forecasts to be unbiased: E[ỹ n (h)] = SPSβ n (h) = Sβ n (h). Revised forecasts are unbiased iff SPS = S. Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 22
55 General properties: variance ỹ n (h) = SPŷ n (h) Let variance of base forecasts ŷ n (h) be given by Σ h = Var[ŷ n (h) y 1,..., y n ] Then the variance of the revised forecasts is given by Var[ỹ n (h) y 1,..., y n ] = SPΣ h P S. Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 23
56 General properties: variance ỹ n (h) = SPŷ n (h) Let variance of base forecasts ŷ n (h) be given by Σ h = Var[ŷ n (h) y 1,..., y n ] Then the variance of the revised forecasts is given by Var[ỹ n (h) y 1,..., y n ] = SPΣ h P S. Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 23
57 General properties: variance ỹ n (h) = SPŷ n (h) Let variance of base forecasts ŷ n (h) be given by Σ h = Var[ŷ n (h) y 1,..., y n ] Then the variance of the revised forecasts is given by Var[ỹ n (h) y 1,..., y n ] = SPΣ h P S. Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 23
58 BLUF via trace minimization Theorem For any P satisfying SPS = S, then min P = trace[spσ h P S ] has solution P = (S Σ h S) 1 S Σ h. Σ h is generalized inverse of Σ h. Revised forecasts ỹ n (h) = S(S Σ h S) 1 S Σ hŷn(h) Base forecasts Equivalent to GLS estimate of regression ŷ n (h) = Sβ n (h) + ε h where ε N(0, Σ h ). Problem: Σ h hard to estimate. Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 24
59 BLUF via trace minimization Theorem For any P satisfying SPS = S, then min P = trace[spσ h P S ] has solution P = (S Σ h S) 1 S Σ h. Σ h is generalized inverse of Σ h. Revised forecasts ỹ n (h) = S(S Σ h S) 1 S Σ hŷn(h) Base forecasts Equivalent to GLS estimate of regression ŷ n (h) = Sβ n (h) + ε h where ε N(0, Σ h ). Problem: Σ h hard to estimate. Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 24
60 BLUF via trace minimization Theorem For any P satisfying SPS = S, then min P = trace[spσ h P S ] has solution P = (S Σ h S) 1 S Σ h. Σ h is generalized inverse of Σ h. Revised forecasts ỹ n (h) = S(S Σ h S) 1 S Σ hŷn(h) Base forecasts Equivalent to GLS estimate of regression ŷ n (h) = Sβ n (h) + ε h where ε N(0, Σ h ). Problem: Σ h hard to estimate. Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 24
61 BLUF via trace minimization Theorem For any P satisfying SPS = S, then min P = trace[spσ h P S ] has solution P = (S Σ h S) 1 S Σ h. Σ h is generalized inverse of Σ h. Revised forecasts ỹ n (h) = S(S Σ h S) 1 S Σ hŷn(h) Base forecasts Equivalent to GLS estimate of regression ŷ n (h) = Sβ n (h) + ε h where ε N(0, Σ h ). Problem: Σ h hard to estimate. Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 24
62 BLUF via trace minimization Theorem For any P satisfying SPS = S, then min P = trace[spσ h P S ] has solution P = (S Σ h S) 1 S Σ h. Σ h is generalized inverse of Σ h. Revised forecasts ỹ n (h) = S(S Σ h S) 1 S Σ hŷn(h) Base forecasts Equivalent to GLS estimate of regression ŷ n (h) = Sβ n (h) + ε h where ε N(0, Σ h ). Problem: Σ h hard to estimate. Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 24
63 BLUF via trace minimization Theorem For any P satisfying SPS = S, then min P = trace[spσ h P S ] has solution P = (S Σ h S) 1 S Σ h. Σ h is generalized inverse of Σ h. Revised forecasts ỹ n (h) = S(S Σ h S) 1 S Σ hŷn(h) Base forecasts Equivalent to GLS estimate of regression ŷ n (h) = Sβ n (h) + ε h where ε N(0, Σ h ). Problem: Σ h hard to estimate. Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 24
64 Optimal combination forecasts Revised forecasts ỹ n (h) = S(S Σ h S) 1 S Σ hŷn(h) Base forecasts Solution 1: OLS Assume ε h Sε B,h where ε B,h is the forecast error at bottom level. Then Σ h SΩ h S where Ω h = Var(ε B,h ). If Moore-Penrose generalized inverse used, then (S Σ h S) 1 S Σ h = (S S) 1 S. ỹ n (h) = S(S S) 1 S ŷ n (h) Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 25
65 Optimal combination forecasts Revised forecasts ỹ n (h) = S(S Σ h S) 1 S Σ hŷn(h) Base forecasts Solution 1: OLS Assume ε h Sε B,h where ε B,h is the forecast error at bottom level. Then Σ h SΩ h S where Ω h = Var(ε B,h ). If Moore-Penrose generalized inverse used, then (S Σ h S) 1 S Σ h = (S S) 1 S. ỹ n (h) = S(S S) 1 S ŷ n (h) Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 25
66 Optimal combination forecasts Revised forecasts ỹ n (h) = S(S Σ h S) 1 S Σ hŷn(h) Base forecasts Solution 1: OLS Assume ε h Sε B,h where ε B,h is the forecast error at bottom level. Then Σ h SΩ h S where Ω h = Var(ε B,h ). If Moore-Penrose generalized inverse used, then (S Σ h S) 1 S Σ h = (S S) 1 S. ỹ n (h) = S(S S) 1 S ŷ n (h) Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 25
67 Optimal combination forecasts Revised forecasts ỹ n (h) = S(S Σ h S) 1 S Σ hŷn(h) Base forecasts Solution 1: OLS Assume ε h Sε B,h where ε B,h is the forecast error at bottom level. Then Σ h SΩ h S where Ω h = Var(ε B,h ). If Moore-Penrose generalized inverse used, then (S Σ h S) 1 S Σ h = (S S) 1 S. ỹ n (h) = S(S S) 1 S ŷ n (h) Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 25
68 Optimal combination forecasts Revised forecasts ỹ n (h) = S(S Σ h S) 1 S Σ hŷn(h) Base forecasts Solution 1: OLS Assume ε h Sε B,h where ε B,h is the forecast error at bottom level. Then Σ h SΩ h S where Ω h = Var(ε B,h ). If Moore-Penrose generalized inverse used, then (S Σ h S) 1 S Σ h = (S S) 1 S. ỹ n (h) = S(S S) 1 S ŷ n (h) Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 25
69 Optimal combination forecasts Revised forecasts ỹ n (h) = S(S Σ h S) 1 S Σ hŷn(h) Base forecasts Solution 1: OLS Assume ε h Sε B,h where ε B,h is the forecast error at bottom level. Then Σ h SΩ h S where Ω h = Var(ε B,h ). If Moore-Penrose generalized inverse used, then (S Σ h S) 1 S Σ h = (S S) 1 S. ỹ n (h) = S(S S) 1 S ŷ n (h) Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 25
70 Optimal combination forecasts Revised forecasts ỹ n (h) = S(S Σ h S) 1 S Σ hŷn(h) Base forecasts Solution 2: WLS Suppose we approximate Σ 1 by its diagonal. Easy to estimate, and places weight where we have best forecasts. Empirically, it gives better forecasts than other available methods. ỹ n (h) = S(S ΛS) 1 S Λŷ n (h) Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 26
71 Optimal combination forecasts Revised forecasts ỹ n (h) = S(S Σ h S) 1 S Σ hŷn(h) Base forecasts Solution 2: WLS Suppose we approximate Σ 1 by its diagonal. Easy to estimate, and places weight where we have best forecasts. Empirically, it gives better forecasts than other available methods. ỹ n (h) = S(S ΛS) 1 S Λŷ n (h) Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 26
72 Optimal combination forecasts Revised forecasts ỹ n (h) = S(S Σ h S) 1 S Σ hŷn(h) Base forecasts Solution 2: WLS Suppose we approximate Σ 1 by its diagonal. Easy to estimate, and places weight where we have best forecasts. Empirically, it gives better forecasts than other available methods. ỹ n (h) = S(S ΛS) 1 S Λŷ n (h) Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 26
73 Optimal combination forecasts Revised forecasts ỹ n (h) = S(S Σ h S) 1 S Σ hŷn(h) Base forecasts Solution 2: WLS Suppose we approximate Σ 1 by its diagonal. Easy to estimate, and places weight where we have best forecasts. Empirically, it gives better forecasts than other available methods. ỹ n (h) = S(S ΛS) 1 S Λŷ n (h) Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 26
74 Optimal combination forecasts Revised forecasts ỹ n (h) = S(S Σ h S) 1 S Σ hŷn(h) Base forecasts Solution 2: WLS Suppose we approximate Σ 1 by its diagonal. Easy to estimate, and places weight where we have best forecasts. Empirically, it gives better forecasts than other available methods. ỹ n (h) = S(S ΛS) 1 S Λŷ n (h) Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 26
75 Optimal combination forecasts Revised forecasts ỹ n (h) = S(S Σ h S) 1 S Σ hŷn(h) Base forecasts Solution 2: WLS Suppose we approximate Σ 1 by its diagonal. Easy to estimate, and places weight where we have best forecasts. Empirically, it gives better forecasts than other available methods. ỹ n (h) = S(S ΛS) 1 S Λŷ n (h) Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 26
76 Challenges ỹ n (h) = S(S ΛS) 1 S Λŷ n (h) Computational difficulties in big hierarchies due to size of the S matrix and singular behavior of (S ΛS). Loss of information in ignoring covariance matrix in computing point forecasts. Still need to estimate covariance matrix to produce prediction intervals. Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 27
77 Challenges ỹ n (h) = S(S ΛS) 1 S Λŷ n (h) Computational difficulties in big hierarchies due to size of the S matrix and singular behavior of (S ΛS). Loss of information in ignoring covariance matrix in computing point forecasts. Still need to estimate covariance matrix to produce prediction intervals. Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 27
78 Challenges ỹ n (h) = S(S ΛS) 1 S Λŷ n (h) Computational difficulties in big hierarchies due to size of the S matrix and singular behavior of (S ΛS). Loss of information in ignoring covariance matrix in computing point forecasts. Still need to estimate covariance matrix to produce prediction intervals. Visualising and forecasting big time series data BLUF: Best Linear Unbiased Forecasts 27
79 Outline 1 Examples of biggish time series 2 Time series visualisation 3 BLUF: Best Linear Unbiased Forecasts 4 Application: Australian tourism 5 Fast computation tricks 6 hts package for R 7 References Visualising and forecasting big time series data Application: Australian tourism 28
80 Australian tourism Visualising and forecasting big time series data Application: Australian tourism 29
81 Australian tourism Hierarchy: States (7) Zones (27) Regions (82) Visualising and forecasting big time series data Application: Australian tourism 29
82 Australian tourism Hierarchy: States (7) Zones (27) Regions (82) Base forecasts ETS (exponential smoothing) models Visualising and forecasting big time series data Application: Australian tourism 29
83 Base forecasts Domestic tourism forecasts: Total Visitor nights Year Visualising and forecasting big time series data Application: Australian tourism 30
84 Base forecasts Domestic tourism forecasts: NSW Visitor nights Year Visualising and forecasting big time series data Application: Australian tourism 30
85 Base forecasts Domestic tourism forecasts: VIC Visitor nights Year Visualising and forecasting big time series data Application: Australian tourism 30
86 Base forecasts Domestic tourism forecasts: Nth.Coast.NSW Visitor nights Year Visualising and forecasting big time series data Application: Australian tourism 30
87 Base forecasts Domestic tourism forecasts: Metro.QLD Visitor nights Year Visualising and forecasting big time series data Application: Australian tourism 30
88 Base forecasts Domestic tourism forecasts: Sth.WA Visitor nights Year Visualising and forecasting big time series data Application: Australian tourism 30
89 Base forecasts Domestic tourism forecasts: X201.Melbourne Visitor nights Year Visualising and forecasting big time series data Application: Australian tourism 30
90 Base forecasts Domestic tourism forecasts: X402.Murraylands Visitor nights Year Visualising and forecasting big time series data Application: Australian tourism 30
91 Base forecasts Domestic tourism forecasts: X809.Daly Visitor nights Year Visualising and forecasting big time series data Application: Australian tourism 30
92 Reconciled forecasts Total Visualising and forecasting big time series data Application: Australian tourism 31
93 Reconciled forecasts NSW VIC QLD Other Visualising and forecasting big time series data Application: Australian tourism 31
94 Reconciled forecasts Sydney Other NSW Melbourne Other VIC GC and Brisbane Other QLD Capital cities Other Visualising and forecasting big time series data Application: Australian tourism 31
95 Forecast evaluation Select models using all observations; Re-estimate models using first 12 observations and generate 1- to 8-step-ahead forecasts; Increase sample size one observation at a time, re-estimate models, generate forecasts until the end of the sample; In total 24 1-step-ahead, 23 2-steps-ahead, up to 17 8-steps-ahead for forecast evaluation. Visualising and forecasting big time series data Application: Australian tourism 32
96 Forecast evaluation Select models using all observations; Re-estimate models using first 12 observations and generate 1- to 8-step-ahead forecasts; Increase sample size one observation at a time, re-estimate models, generate forecasts until the end of the sample; In total 24 1-step-ahead, 23 2-steps-ahead, up to 17 8-steps-ahead for forecast evaluation. Visualising and forecasting big time series data Application: Australian tourism 32
97 Forecast evaluation Select models using all observations; Re-estimate models using first 12 observations and generate 1- to 8-step-ahead forecasts; Increase sample size one observation at a time, re-estimate models, generate forecasts until the end of the sample; In total 24 1-step-ahead, 23 2-steps-ahead, up to 17 8-steps-ahead for forecast evaluation. Visualising and forecasting big time series data Application: Australian tourism 32
98 Forecast evaluation Select models using all observations; Re-estimate models using first 12 observations and generate 1- to 8-step-ahead forecasts; Increase sample size one observation at a time, re-estimate models, generate forecasts until the end of the sample; In total 24 1-step-ahead, 23 2-steps-ahead, up to 17 8-steps-ahead for forecast evaluation. Visualising and forecasting big time series data Application: Australian tourism 32
99 Hierarchy: states, zones, regions MAPE h = 1 h = 2 h = 4 h = 6 h = 8 Average Top Level: Australia Bottom-up OLS WLS Level: States Bottom-up OLS WLS Level: Zones Bottom-up OLS WLS Bottom Level: Regions Bottom-up OLS WLS Visualising and forecasting big time series data Application: Australian tourism 33
100 Outline 1 Examples of biggish time series 2 Time series visualisation 3 BLUF: Best Linear Unbiased Forecasts 4 Application: Australian tourism 5 Fast computation tricks 6 hts package for R 7 References Visualising and forecasting big time series data Fast computation tricks 34
101 Fast computation: hierarchical data Total A B C y t = AX AY AZ Y t Y A,t Y B,t Y C,t Y AX,t Y AY,t Y AZ,t Y BX,t Y BY,t Y BZ,t Y CX,t Y CY,t Y CZ,t = BX BY BZ } {{ } S Y AX,t Y AY,t Y AZ,t Y BX,t Y BY,t Y BZ,t Y CX,t Y CY,t Y CZ,t }{{} B t CX CY CZ y t = SB t Visualising and forecasting big time series data Fast computation tricks 35
102 Fast computation: hierarchical data Total A B C y t = AX AY AZ Y t Y A,t Y AX,t Y AY,t Y AZ,t Y B,t Y BX,t Y BY,t Y BZ,t Y C,t Y CX,t Y CY,t Y CZ,t = BX BY BZ } {{ } S Y AX,t Y AY,t Y AZ,t Y BX,t Y BY,t Y BZ,t Y CX,t Y CY,t Y CZ,t }{{} B t CX CY CZ y t = SB t Visualising and forecasting big time series data Fast computation tricks 36
103 Fast computation: hierarchies Think of the hierarchy as a tree of trees: Total T 1 T 2... T K Then the summing matrix contains k smaller summing matrices: S = 1 n 1 1 n 2 1 n K S S S K where 1 n is an n-vector of ones and tree T i has n i terminal nodes. Visualising and forecasting big time series data Fast computation tricks 37.
104 Fast computation: hierarchies Think of the hierarchy as a tree of trees: Total T 1 T 2... T K Then the summing matrix contains k smaller summing matrices: S = 1 n 1 1 n 2 1 n K S S S K where 1 n is an n-vector of ones and tree T i has n i terminal nodes. Visualising and forecasting big time series data Fast computation tricks 37.
105 Fast computation: hierarchies S ΛS = S 1Λ 1 S S 2Λ 2 S S KΛ K S K. +λ 0 J n λ 0 is the top left element of Λ; Λ k is a block of Λ, corresponding to tree T k ; J n is a matrix of ones; n = k n k. Now apply the Sherman-Morrison formula... Visualising and forecasting big time series data Fast computation tricks 38
106 Fast computation: hierarchies S ΛS = S 1Λ 1 S S 2Λ 2 S S KΛ K S K. +λ 0 J n λ 0 is the top left element of Λ; Λ k is a block of Λ, corresponding to tree T k ; J n is a matrix of ones; n = k n k. Now apply the Sherman-Morrison formula... Visualising and forecasting big time series data Fast computation tricks 38
107 Fast computation: hierarchies (S 1 Λ 1S 1 ) (S ΛS) 1 0 (S 2 = Λ 2S 2 ) cs (S K Λ KS K ) 1 S 0 can be partitioned into K 2 blocks, with the (k, l) block (of dimension n k n l ) being J nk,n l (S k Λ ks k ) 1 J nk,n l (S l Λ ls l ) 1 is a n k n l matrix of ones. c 1 = λ n k (S k Λ ks k ) 1 1 nk. k Each S k Λ ks k can be inverted similarly. S Λy can also be computed recursively. Visualising and forecasting big time series data Fast computation tricks 39
108 Fast computation: hierarchies (S 1 Λ 1S 1 ) (S ΛS) 1 0 (S 2 = Λ 2S 2 ) cs (S K Λ KS K ) 1 S 0 can be partitioned into K 2 blocks, with the (k, l) block (of dimension n k n l ) being The recursive (S k Λ calculations ks k ) 1 J nk,n l (S l Λ can ls l ) 1 be done in such a way that we never J nk,n l store is a n k any n l matrix of ones. of the large matrices c 1 = involved. λ n k (S k Λ ks k ) 1 1 nk. k Each S k Λ ks k can be inverted similarly. S Λy can also be computed recursively. Visualising and forecasting big time series data Fast computation tricks 39
109 Fast computation A similar algorithm has been developed for grouped time series with two groups. When the time series are not strictly hierarchical and have more than two grouping variables: Use sparse matrix storage and arithmetic. Use iterative approximation for inverting large sparse matrices. Paige & Saunders (1982) ACM Trans. Math. Software Visualising and forecasting big time series data Fast computation tricks 40
110 Fast computation A similar algorithm has been developed for grouped time series with two groups. When the time series are not strictly hierarchical and have more than two grouping variables: Use sparse matrix storage and arithmetic. Use iterative approximation for inverting large sparse matrices. Paige & Saunders (1982) ACM Trans. Math. Software Visualising and forecasting big time series data Fast computation tricks 40
111 Fast computation A similar algorithm has been developed for grouped time series with two groups. When the time series are not strictly hierarchical and have more than two grouping variables: Use sparse matrix storage and arithmetic. Use iterative approximation for inverting large sparse matrices. Paige & Saunders (1982) ACM Trans. Math. Software Visualising and forecasting big time series data Fast computation tricks 40
112 Outline 1 Examples of biggish time series 2 Time series visualisation 3 BLUF: Best Linear Unbiased Forecasts 4 Application: Australian tourism 5 Fast computation tricks 6 hts package for R 7 References Visualising and forecasting big time series data hts package for R 41
113 hts package for R hts: Hierarchical and grouped time series Methods for analysing and forecasting hierarchical and grouped time series Version: 4.5 Depends: forecast ( 5.0), SparseM Imports: parallel, utils Published: Author: Rob J Hyndman, Earo Wang and Alan Lee Maintainer: Rob J Hyndman <Rob.Hyndman at monash.edu> BugReports: License: GPL ( 2) Visualising and forecasting big time series data hts package for R 42
114 Example using R library(hts) # bts is a matrix containing the bottom level time series # nodes describes the hierarchical structure y <- hts(bts, nodes=list(2, c(3,2))) Visualising and forecasting big time series data hts package for R 43
115 Example using R library(hts) # bts is a matrix containing the bottom level time series # nodes describes the hierarchical structure y <- hts(bts, nodes=list(2, c(3,2))) Total A B AX AY AZ BX BY Visualising and forecasting big time series data hts package for R 43
116 Example using R library(hts) # bts is a matrix containing the bottom level time series # nodes describes the hierarchical structure y <- hts(bts, nodes=list(2, c(3,2))) # Forecast 10-step-ahead using WLS combination method # ETS used for each series by default fc <- forecast(y, h=10) Visualising and forecasting big time series data hts package for R 44
117 forecast.gts function Usage forecast(object, h, method = c("comb", "bu", "mo", "tdgsf", "tdgsa", "tdfp"), fmethod = c("ets", "rw", "arima"), weights = c("sd", "none", "nseries"), positive = FALSE, parallel = FALSE, num.cores = 2,...) Arguments object Hierarchical time series object of class gts. h Forecast horizon method Method for distributing forecasts within the hierarchy. fmethod Forecasting method to use positive If TRUE, forecasts are forced to be strictly positive weights Weights used for "optimal combination" method. When weights = "sd", it takes account of the standard deviation of forecasts. parallel If TRUE, allow parallel processing num.cores If parallel = TRUE, specify how many cores are going to be used Visualising and forecasting big time series data hts package for R 45
118 Outline 1 Examples of biggish time series 2 Time series visualisation 3 BLUF: Best Linear Unbiased Forecasts 4 Application: Australian tourism 5 Fast computation tricks 6 hts package for R 7 References Visualising and forecasting big time series data References 46
119 References RJ Hyndman, RA Ahmed, G Athanasopoulos, and HL Shang (2011). Optimal combination forecasts for hierarchical time series. Computational statistics & data analysis 55(9), RJ Hyndman, AJ Lee, and E Wang (2014). Fast computation of reconciled forecasts for hierarchical and grouped time series. Working paper 17/14. Department of Econometrics & Business Statistics, Monash University RJ Hyndman, AJ Lee, and E Wang (2014). hts: Hierarchical and grouped time series. cran.r-project.org/package=hts. RJ Hyndman and G Athanasopoulos (2014). Forecasting: principles and practice. OTexts. OTexts.org/fpp/. Visualising and forecasting big time series data References 47
120 References RJ Hyndman, RA Ahmed, G Athanasopoulos, and HL Shang (2011). Optimal combination forecasts for hierarchical time series. Computational statistics & data analysis 55(9), RJ Hyndman, AJ Lee, and E Wang (2014). Fast computation of reconciled forecasts for hierarchical and grouped time series. Working paper 17/14. Department of Econometrics & Business Statistics, Monash University RJ Hyndman, AJ Lee, and E Wang (2014). hts: Hierarchical and grouped time series. cran.r-project.org/package=hts. RJ robjhyndman.com Hyndman and G Athanasopoulos (2014). Forecasting: principles and practice. OTexts. OTexts.org/fpp/. Papers and R code: Rob.Hyndman@monash.edu Visualising and forecasting big time series data References 47
Fast computation of reconciled forecasts in hierarchical and grouped time series
Rob J Hyndman Fast computation of reconciled forecasts in hierarchical and grouped time series Total A B C AX AY AZ BX BY BZ CX CY CZ 1 Outline 1 Optimally reconciled forecasts 2 Fast computation 3 hts
More informationForecasting: principles and practice. Rob J Hyndman 4 Hierarchical forecasting
Forecasting: principles and practice Rob J Hyndman 4 Hierarchical forecasting 1 Outline 1 Hierarchical and grouped time series 2 Optimal forecast reconciliation 3 hts package for R 4 Application: Australian
More informationOptimal forecast reconciliation for big time series data
Rob J Hyndman Optimal forecast reconciliation for big time series data Total A AA AB AC B BA BB BC C CA CB CC Follow along using R Requirements Install the hts package and its dependencies. Optimal forecast
More informationForecasting hierarchical (and grouped) time series
George Athanasopoulos (with Rob J. Hyndman, Nikos Kourentzes and Fotis Petropoulos) Forecasting hierarchical (and grouped) time series Total A B C AA AB AC BA BB BC CA CB CC 1 Outline 1 Hierarchical and
More informationOptimal combination forecasts for hierarchical time series
Optimal combination forecasts for hierarchical time series Rob J. Hyndman Roman A. Ahmed Department of Econometrics and Business Statistics Outline 1 Review of hierarchical forecasting 2 3 Simulation study
More informationFast computation of reconciled forecasts for hierarchical and grouped time series
ISSN 1440-771X Department of Econometrics and Business Statistics http://wwwbusecomonasheduau/depts/ebs/pubs/wpapers/ Fast computation of reconciled forecasts for hierarchical and grouped time series Rob
More informationTime series can often be naturally disaggregated
Optimally Reconciling Forecasts in a Hierarchy Rob J. Hyndman and George Athanasopoulos Preview We know that when forecasting the product hierarchy, reconciliation is invariably needed to make the sum
More informationForecasting in Hierarchical Models
Forecasting in Hierarchical Models Lucy Morgan Supervisor: Nikolaos Kourentzes 20 th February 2015 Introduction Forecasting is the process of making statements about events whose actual outcomes (typically)
More informationOptimal forecast reconciliation for hierarchical and grouped time series through trace minimization
ISSN 1440-771X Department of Econometrics and Business Statistics http://business.monash.edu/econometrics-and-businessstatistics/research/publications Optimal forecast reconciliation for hierarchical and
More informationOptimal forecast reconciliation for hierarchical and grouped time series through trace minimization
Optimal forecast reconciliation for hierarchical and grouped time series through trace minimization Shanika L Wickramasuriya Department of Econometrics and Business Statistics, Monash University, VIC 3800,
More informationModelling and Forecasting Australian Domestic Tourism
Modelling and Forecasting Australian Domestic Tourism George Athanasopoulos & Rob Hyndman Outline Background 1 Background 2 Data 3 Regression models 4 Exponential smoothing via innovations state space
More informationGrouped time-series forecasting: Application to regional infant mortality counts
Grouped time-series forecasting: Application to regional infant mortality counts Han Lin Shang and Peter W. F. Smith University of Southampton Motivation 1 Multiple time series can be disaggregated by
More informationForecasting hierarchical and grouped time series through trace minimization
ISSN 1440-771X Department of Econometrics and Business Statistics http://business.monash.edu/econometrics-and-business-statistics/research/publications Forecasting hierarchical and grouped time series
More informationPackage thief. R topics documented: January 24, Version 0.3 Title Temporal Hierarchical Forecasting
Version 0.3 Title Temporal Hierarchical Forecasting Package thief January 24, 2018 Methods and tools for generating forecasts at different temporal frequencies using a hierarchical time series approach.
More informationForecast UPC-Level FMCG Demand, Part III: Grouped Reconciliation
2016 IEEE International Conference on Big Data (Big Data) Forecast UPC-Level FMCG Demand, Part III: Grouped Reconciliation Dazhi Yang, Gary S. W. Goh, Siwei Jiang and Allan N. Zhang Singapore Institute
More informationForecasting: Principles and Practice. Rob J Hyndman. 1. Introduction to forecasting OTexts.org/fpp/1/ OTexts.org/fpp/2/3
Rob J Hyndman Forecasting: Principles and Practice 1. Introduction to forecasting OTexts.org/fpp/1/ OTexts.org/fpp/2/3 Forecasting: Principles and Practice 1 Resources Slides Exercises Textbook Useful
More informationForecasting: principles and practice. Rob J Hyndman 1.1 Introduction to Forecasting
Forecasting: principles and practice Rob J Hyndman 1.1 Introduction to Forecasting 1 Outline 1 Background 2 Case studies 3 The statistical forecasting perspective 4 What can we forecast? 2 Resources Slides
More informationUsing Temporal Hierarchies to Predict Tourism Demand
Using Temporal Hierarchies to Predict Tourism Demand International Symposium on Forecasting 29 th June 2015 Nikolaos Kourentzes Lancaster University George Athanasopoulos Monash University n.kourentzes@lancaster.ac.uk
More informationForecasting 1 to h steps ahead using partial least squares
Forecasting 1 to h steps ahead using partial least squares Philip Hans Franses Econometric Institute, Erasmus University Rotterdam November 10, 2006 Econometric Institute Report 2006-47 I thank Dick van
More informationThe value of feedback in forecasting competitions
ISSN 1440-771X Department of Econometrics and Business Statistics http://www.buseco.monash.edu.au/depts/ebs/pubs/wpapers/ The value of feedback in forecasting competitions George Athanasopoulos and Rob
More informationDEPARTMENT OF ECONOMETRICS AND BUSINESS STATISTICS
ISSN 1440-771X ISBN 0 7326 1085 0 Unmasking the Theta Method Rob J. Hyndman and Baki Billah Working Paper 5/2001 2001 DEPARTMENT OF ECONOMETRICS AND BUSINESS STATISTICS AUSTRALIA Unmasking the Theta method
More informationModelling and forecasting Australian domestic tourism
* Title page with author details Modelling and forecasting Australian domestic tourism G. Athanasopoulos, R. J. Hyndman Department of Econometrics and Business Statistics, Monash University, Melbourne,
More informationA tale of two cities. John Daley, CEO, Grattan Institute Work and life in cities: City strategy in Australia Melbourne Economic Forum 27 October 2016
A tale of two cities John Daley, CEO, Grattan Institute Work and life in cities: City strategy in Australia Melbourne Economic Forum 27 October 2016 A tale of two cities Shifts in consumption are driving
More informationComponents for Accurate Forecasting & Continuous Forecast Improvement
Components for Accurate Forecasting & Continuous Forecast Improvement An ISIS Solutions White Paper November 2009 Page 1 Achieving forecast accuracy for business applications one year in advance requires
More informationCross-temporal coherent forecasts for Australian tourism
ISSN 1440-771X Department of Econometrics and Business Statistics http://business.monash.edu/econometrics-and-businessstatistics/research/publications Cross-temporal coherent forecasts for Australian tourism
More informationAutomatic forecasting with a modified exponential smoothing state space framework
ISSN 1440-771X Department of Econometrics and Business Statistics http://www.buseco.monash.edu.au/depts/ebs/pubs/wpapers/ Automatic forecasting with a modified exponential smoothing state space framework
More informationDepartment of Econometrics and Business Statistics
ISSN 1440-771X Australia Department of Econometrics and Business Statistics http://www.buseco.monash.edu.au/depts/ebs/pubs/wpapers/ Exponential Smoothing: A Prediction Error Decomposition Principle Ralph
More informationECON 427: ECONOMIC FORECASTING. Ch1. Getting started OTexts.org/fpp2/
ECON 427: ECONOMIC FORECASTING Ch1. Getting started OTexts.org/fpp2/ 1 Outline 1 What can we forecast? 2 Time series data 3 Some case studies 4 The statistical forecasting perspective 2 Forecasting is
More informationA MACRO-DRIVEN FORECASTING SYSTEM FOR EVALUATING FORECAST MODEL PERFORMANCE
A MACRO-DRIVEN ING SYSTEM FOR EVALUATING MODEL PERFORMANCE Bryan Sellers Ross Laboratories INTRODUCTION A major problem of forecasting aside from obtaining accurate forecasts is choosing among a wide range
More informationLecture 1: Introduction to Forecasting
NATCOR: Forecasting & Predictive Analytics Lecture 1: Introduction to Forecasting Professor John Boylan Lancaster Centre for Forecasting Department of Management Science Leading research centre in applied
More informationRegularization in Hierarchical Time Series Forecasting With Application to Electricity Smart Meter Data
Regularization in Hierarchical Time Series Forecasting With Application to Electricity Smart Meter Data Souhaib Ben Taieb Monash Business School Monash University Jiafan Yu Dept. of CEE Stanford University
More informationTime Series Outlier Detection
Time Series Outlier Detection Tingyi Zhu July 28, 2016 Tingyi Zhu Time Series Outlier Detection July 28, 2016 1 / 42 Outline Time Series Basics Outliers Detection in Single Time Series Outlier Series Detection
More informationMeta-learning how to forecast time series
ISSN 1440-771X Department of Econometrics and Business Statistics http://business.monash.edu/econometrics-and-businessstatistics/research/publications Meta-learning how to forecast time series Thiyanga
More informationRegression of Time Series
Mahlerʼs Guide to Regression of Time Series CAS Exam S prepared by Howard C. Mahler, FCAS Copyright 2016 by Howard C. Mahler. Study Aid 2016F-S-9Supplement Howard Mahler hmahler@mac.com www.howardmahler.com/teaching
More informationFunctional time series
Rob J Hyndman Functional time series with applications in demography 4. Connections, extensions and applications Outline 1 Yield curves 2 Electricity prices 3 Dynamic updating with partially observed functions
More informationProduct and Inventory Management (35E00300) Forecasting Models Trend analysis
Product and Inventory Management (35E00300) Forecasting Models Trend analysis Exponential Smoothing Data Storage Shed Sales Period Actual Value(Y t ) Ŷ t-1 α Y t-1 Ŷ t-1 Ŷ t January 10 = 10 0.1 February
More informationLecture 4 Forecasting
King Saud University College of Computer & Information Sciences IS 466 Decision Support Systems Lecture 4 Forecasting Dr. Mourad YKHLEF The slides content is derived and adopted from many references Outline
More informationA new approach to forecasting based on exponential smoothing with independent regressors
ISSN 1440-771X Australia Department of Econometrics and Business Statistics http://wwwbusecomonasheduau/depts/ebs/pubs/wpapers/ A new approach to forecasting based on exponential smoothing with independent
More informationThe SAB Medium Term Sales Forecasting System : From Data to Planning Information. Kenneth Carden SAB : Beer Division Planning
The SAB Medium Term Sales Forecasting System : From Data to Planning Information Kenneth Carden SAB : Beer Division Planning Planning in Beer Division F Operational planning = what, when, where & how F
More informationDEPARTMENT OF ECONOMETRICS AND BUSINESS STATISTICS
ISSN 1440-771X ISBN 0 7326 1091 5 Prediction Intervals for Exponential Smoothing State Space Models Rob J Hyndman, Anne B Koehler, J. Keith Ord and Ralph D Snyder Working Paper 11/2001 2001 DEPARTMENT
More informationZhaoxing Gao and Ruey S Tsay Booth School of Business, University of Chicago. August 23, 2018
Supplementary Material for Structural-Factor Modeling of High-Dimensional Time Series: Another Look at Approximate Factor Models with Diverging Eigenvalues Zhaoxing Gao and Ruey S Tsay Booth School of
More informationNonparametric time series forecasting with dynamic updating
18 th World IMAS/MODSIM Congress, Cairns, Australia 13-17 July 2009 http://mssanz.org.au/modsim09 1 Nonparametric time series forecasting with dynamic updating Han Lin Shang and Rob. J. Hyndman Department
More informationMathematics I. Exercises with solutions. 1 Linear Algebra. Vectors and Matrices Let , C = , B = A = Determine the following matrices:
Mathematics I Exercises with solutions Linear Algebra Vectors and Matrices.. Let A = 5, B = Determine the following matrices: 4 5, C = a) A + B; b) A B; c) AB; d) BA; e) (AB)C; f) A(BC) Solution: 4 5 a)
More informationARMA (and ARIMA) models are often expressed in backshift notation.
Backshift Notation ARMA (and ARIMA) models are often expressed in backshift notation. B is the backshift operator (also called the lag operator ). It operates on time series, and means back up by one time
More informationSimultaneous Equation Models (Book Chapter 5)
Simultaneous Equation Models (Book Chapter 5) Interrelated equations with continuous dependent variables: Utilization of individual vehicles (measured in kilometers driven) in multivehicle households Interrelation
More informationPhD/MA Econometrics Examination January 2012 PART A
PhD/MA Econometrics Examination January 2012 PART A ANSWER ANY TWO QUESTIONS IN THIS SECTION NOTE: (1) The indicator function has the properties: (2) Question 1 Let, [defined as if using the indicator
More informationRob J Hyndman. Forecasting using. 3. Autocorrelation and seasonality OTexts.com/fpp/2/ OTexts.com/fpp/6/1. Forecasting using R 1
Rob J Hyndman Forecasting using 3. Autocorrelation and seasonality OTexts.com/fpp/2/ OTexts.com/fpp/6/1 Forecasting using R 1 Outline 1 Time series graphics 2 Seasonal or cyclic? 3 Autocorrelation Forecasting
More informationChapter 2: Matrix Algebra
Chapter 2: Matrix Algebra (Last Updated: October 12, 2016) These notes are derived primarily from Linear Algebra and its applications by David Lay (4ed). Write A = 1. Matrix operations [a 1 a n. Then entry
More informationForecasting using R. Rob J Hyndman. 2.3 Stationarity and differencing. Forecasting using R 1
Forecasting using R Rob J Hyndman 2.3 Stationarity and differencing Forecasting using R 1 Outline 1 Stationarity 2 Differencing 3 Unit root tests 4 Lab session 10 5 Backshift notation Forecasting using
More informationPackage Tcomp. June 6, 2018
Package Tcomp June 6, 2018 Title Data from the 2010 Tourism Forecasting Competition Version 1.0.1 The 1311 time series from the tourism forecasting competition conducted in 2010 and described in Athanasopoulos
More informationThe Art of Forecasting
Time Series The Art of Forecasting Learning Objectives Describe what forecasting is Explain time series & its components Smooth a data series Moving average Exponential smoothing Forecast using trend models
More informationS-GSTAR-SUR Model for Seasonal Spatio Temporal Data Forecasting ABSTRACT
Malaysian Journal of Mathematical Sciences (S) March : 53-65 (26) Special Issue: The th IMT-GT International Conference on Mathematics, Statistics and its Applications 24 (ICMSA 24) MALAYSIAN JOURNAL OF
More informationReliability and Risk Analysis. Time Series, Types of Trend Functions and Estimates of Trends
Reliability and Risk Analysis Stochastic process The sequence of random variables {Y t, t = 0, ±1, ±2 } is called the stochastic process The mean function of a stochastic process {Y t} is the function
More informationExercises - Time series analysis
Descriptive analysis of a time series (1) Estimate the trend of the series of gasoline consumption in Spain using a straight line in the period from 1945 to 1995 and generate forecasts for 24 months. Compare
More informationForecasting. Operations Analysis and Improvement Spring
Forecasting Operations Analysis and Improvement 2015 Spring Dr. Tai-Yue Wang Industrial and Information Management Department National Cheng Kung University 1-2 Outline Introduction to Forecasting Subjective
More informationVisualising Forecasting Algorithm Performance using Time Series Instance Spaces
ISSN 44-77X Department of Econometrics and Business Statistics http://business.monash.edu/econometrics-and-business-statistics/research/publications Visualising Forecasting Algorithm Performance using
More informationLecture 7a: Vector Autoregression (VAR)
Lecture 7a: Vector Autoregression (VAR) 1 2 Big Picture We are done with univariate time series analysis Now we switch to multivariate analysis, that is, studying several time series simultaneously. VAR
More informationLecture 7a: Vector Autoregression (VAR)
Lecture 7a: Vector Autoregression (VAR) 1 Big Picture We are done with univariate time series analysis Now we switch to multivariate analysis, that is, studying several time series simultaneously. VAR
More informationMatrix operations Linear Algebra with Computer Science Application
Linear Algebra with Computer Science Application February 14, 2018 1 Matrix operations 11 Matrix operations If A is an m n matrix that is, a matrix with m rows and n columns then the scalar entry in the
More informationVectors and Matrices Statistics with Vectors and Matrices
Vectors and Matrices Statistics with Vectors and Matrices Lecture 3 September 7, 005 Analysis Lecture #3-9/7/005 Slide 1 of 55 Today s Lecture Vectors and Matrices (Supplement A - augmented with SAS proc
More informationAnalisi Statistica per le Imprese
, Analisi Statistica per le Imprese Dip. di Economia Politica e Statistica 4.3. 1 / 33 You should be able to:, Underst model building using multiple regression analysis Apply multiple regression analysis
More informationMa 3/103: Lecture 25 Linear Regression II: Hypothesis Testing and ANOVA
Ma 3/103: Lecture 25 Linear Regression II: Hypothesis Testing and ANOVA March 6, 2017 KC Border Linear Regression II March 6, 2017 1 / 44 1 OLS estimator 2 Restricted regression 3 Errors in variables 4
More informationForecasting: Methods and Applications
Neapolis University HEPHAESTUS Repository School of Economic Sciences and Business http://hephaestus.nup.ac.cy Books 1998 Forecasting: Methods and Applications Makridakis, Spyros John Wiley & Sons, Inc.
More informationFinal A. Problem Points Score Total 100. Math115A Nadja Hempel 03/23/2017
Final A Math115A Nadja Hempel 03/23/2017 nadja@math.ucla.edu Name: UID: Problem Points Score 1 10 2 20 3 5 4 5 5 9 6 5 7 7 8 13 9 16 10 10 Total 100 1 2 Exercise 1. (10pt) Let T : V V be a linear transformation.
More informationLecture: Simultaneous Equation Model (Wooldridge s Book Chapter 16)
Lecture: Simultaneous Equation Model (Wooldridge s Book Chapter 16) 1 2 Model Consider a system of two regressions y 1 = β 1 y 2 + u 1 (1) y 2 = β 2 y 1 + u 2 (2) This is a simultaneous equation model
More informationForecasting: principles and practice 1
Forecasting: principles and practice Rob J Hyndman 2.3 Stationarity and differencing Forecasting: principles and practice 1 Outline 1 Stationarity 2 Differencing 3 Unit root tests 4 Lab session 10 5 Backshift
More informationTheory of Computation - Module 3
Theory of Computation - Module 3 Syllabus Context Free Grammar Simplification of CFG- Normal forms-chomsky Normal form and Greibach Normal formpumping lemma for Context free languages- Applications of
More informationExponentially weighted forecasts
FORECASTING USING R Exponentially weighted forecasts Rob Hyndman Author, forecast Simple exponential smoothing Forecasting Notation: ŷ t+h t = y t+h y 1,...,y t point forecast of Forecast Equation: given
More informationWIND INTEGRATION IN ELECTRICITY GRIDS WORK PACKAGE 3: SIMULATION USING HISTORICAL WIND DATA
WIND INTEGRATION IN ELECTRICITY GRIDS WORK PACKAGE 3: SIMULATION USING PREPARED BY: Strategy and Economics DATE: 18 January 2012 FINAL Australian Energy Market Operator Ltd ABN 94 072 010 327 www.aemo.com.au
More informationProblem Set 2 Solution Sketches Time Series Analysis Spring 2010
Problem Set 2 Solution Sketches Time Series Analysis Spring 2010 Forecasting 1. Let X and Y be two random variables such that E(X 2 ) < and E(Y 2 )
More informationLocal linear forecasts using cubic smoothing splines
Local linear forecasts using cubic smoothing splines Rob J. Hyndman, Maxwell L. King, Ivet Pitrun, Baki Billah 13 January 2004 Abstract: We show how cubic smoothing splines fitted to univariate time series
More informationA time series is called strictly stationary if the joint distribution of every collection (Y t
5 Time series A time series is a set of observations recorded over time. You can think for example at the GDP of a country over the years (or quarters) or the hourly measurements of temperature over a
More informationMathematical Focus 1 Complex numbers adhere to certain arithmetic properties for which they and their complex conjugates are defined.
Situation: Complex Roots in Conjugate Pairs Prepared at University of Georgia Center for Proficiency in Teaching Mathematics June 30, 2013 Sarah Major Prompt: A teacher in a high school Algebra class has
More informationVCE Further Mathematics Units 3&4
Trial Examination 2016 VCE Further Mathematics Units 3&4 Written Examination 2 Question and Answer Booklet Reading time: 15 minutes Writing time: 1 hour 30 minutes Student s Name: Teacher s Name: Structure
More informationU.C. Berkeley CS270: Algorithms Lecture 21 Professor Vazirani and Professor Rao Last revised. Lecture 21
U.C. Berkeley CS270: Algorithms Lecture 21 Professor Vazirani and Professor Rao Scribe: Anupam Last revised Lecture 21 1 Laplacian systems in nearly linear time Building upon the ideas introduced in the
More informationInverse of a Square Matrix. For an N N square matrix A, the inverse of A, 1
Inverse of a Square Matrix For an N N square matrix A, the inverse of A, 1 A, exists if and only if A is of full rank, i.e., if and only if no column of A is a linear combination 1 of the others. A is
More informationANSWERS CHAPTER 15 THINK IT OVER EXERCISES. Nick Lee and Mike Peters think it over. No answers required.
ANSWERS CHAPTER 15 THINK IT OVER think it over No answers required. EXERCISES 1. (a) Yes it could but you would have to be careful in interpreting the results. Transformers tend to be situated outside
More informationSTAT 100C: Linear models
STAT 100C: Linear models Arash A. Amini June 9, 2018 1 / 56 Table of Contents Multiple linear regression Linear model setup Estimation of β Geometric interpretation Estimation of σ 2 Hat matrix Gram matrix
More informationPrincipal Component Analysis and Linear Discriminant Analysis
Principal Component Analysis and Linear Discriminant Analysis Ying Wu Electrical Engineering and Computer Science Northwestern University Evanston, IL 60208 http://www.eecs.northwestern.edu/~yingwu 1/29
More informationForecasting with R A practical workshop
Forecasting with R A practical workshop International Symposium on Forecasting 2016 19 th June 2016 Nikolaos Kourentzes nikolaos@kourentzes.com http://nikolaos.kourentzes.com Fotios Petropoulos fotpetr@gmail.com
More informationAssessing the Employment Agglomeration and Social Accessibility Impacts of High Speed Rail in Eastern Australia: Sydney-Canberra-Melbourne Corridor
Assessing the Employment Agglomeration and Social Accessibility Impacts of High Speed Rail in Eastern Australia: Sydney-Canberra-Melbourne Corridor Professor David A. Hensher FASSA Founding Director Institute
More informationCPSC 313 Introduction to Computability
CPSC 313 Introduction to Computability Grammars in Chomsky Normal Form (Cont d) (Sipser, pages 109-111 (3 rd ed) and 107-109 (2 nd ed)) Renate Scheidler Fall 2018 Chomsky Normal Form A context-free grammar
More informationLinear Programming The Simplex Algorithm: Part II Chapter 5
1 Linear Programming The Simplex Algorithm: Part II Chapter 5 University of Chicago Booth School of Business Kipp Martin October 17, 2017 Outline List of Files Key Concepts Revised Simplex Revised Simplex
More informationForecasting Seasonal Time Series 1. Introduction. Philip Hans Franses Econometric Institute Erasmus University Rotterdam
Forecasting Seasonal Time Series 1. Introduction Philip Hans Franses Econometric Institute Erasmus University Rotterdam SMU and NUS, Singapore, April-May 2004 1 Outline of tutorial lectures 1 Introduction
More informationMath 118, Fall 2014 Final Exam
Math 8, Fall 4 Final Exam True or false Please circle your choice; no explanation is necessary True There is a linear transformation T such that T e ) = e and T e ) = e Solution Since T is linear, if T
More informationftsa: An R Package for Analyzing Functional Time Series by Han Lin Shang
CONTRIBUTED RESEARCH ARTICLES 64 ftsa: An R Package for Analyzing Functional Time Series by Han Lin Shang Abstract Recent advances in computer recording and storing technology have tremendously increased
More informationECON 4160, Spring term 2015 Lecture 7
ECON 4160, Spring term 2015 Lecture 7 Identification and estimation of SEMs (Part 1) Ragnar Nymoen Department of Economics 8 Oct 2015 1 / 55 HN Ch 15 References to Davidson and MacKinnon, Ch 8.1-8.5 Ch
More informationHomoskedasticity. Var (u X) = σ 2. (23)
Homoskedasticity How big is the difference between the OLS estimator and the true parameter? To answer this question, we make an additional assumption called homoskedasticity: Var (u X) = σ 2. (23) This
More informationComputing least squares condition numbers on hybrid multicore/gpu systems
Computing least squares condition numbers on hybrid multicore/gpu systems M. Baboulin and J. Dongarra and R. Lacroix Abstract This paper presents an efficient computation for least squares conditioning
More informationSketching as a Tool for Numerical Linear Algebra
Sketching as a Tool for Numerical Linear Algebra (Part 2) David P. Woodruff presented by Sepehr Assadi o(n) Big Data Reading Group University of Pennsylvania February, 2015 Sepehr Assadi (Penn) Sketching
More informationECO 513 Fall 2009 C. Sims CONDITIONAL EXPECTATION; STOCHASTIC PROCESSES
ECO 513 Fall 2009 C. Sims CONDIIONAL EXPECAION; SOCHASIC PROCESSES 1. HREE EXAMPLES OF SOCHASIC PROCESSES (I) X t has three possible time paths. With probability.5 X t t, with probability.25 X t t, and
More informationDennis Bricker Dept of Mechanical & Industrial Engineering The University of Iowa. Forecasting demand 02/06/03 page 1 of 34
demand -5-4 -3-2 -1 0 1 2 3 Dennis Bricker Dept of Mechanical & Industrial Engineering The University of Iowa Forecasting demand 02/06/03 page 1 of 34 Forecasting is very difficult. especially about the
More informationA Hybrid Method of Forecasting in the Case of the Average Daily Number of Patients
Journal of Computations & Modelling, vol.4, no.3, 04, 43-64 ISSN: 79-765 (print), 79-8850 (online) Scienpress Ltd, 04 A Hybrid Method of Forecasting in the Case of the Average Daily Number of Patients
More informationDiploma Part 2. Quantitative Methods. Examiner s Suggested Answers
Diploma Part Quantitative Methods Examiner s Suggested Answers Question 1 (a) The standard normal distribution has a symmetrical and bell-shaped graph with a mean of zero and a standard deviation equal
More informationChart types and when to use them
APPENDIX A Chart types and when to use them Pie chart Figure illustration of pie chart 2.3 % 4.5 % Browser Usage for April 2012 18.3 % 38.3 % Internet Explorer Firefox Chrome Safari Opera 35.8 % Pie chart
More informationSTAT 212 Business Statistics II 1
STAT 1 Business Statistics II 1 KING FAHD UNIVERSITY OF PETROLEUM & MINERALS DEPARTMENT OF MATHEMATICAL SCIENCES DHAHRAN, SAUDI ARABIA STAT 1: BUSINESS STATISTICS II Semester 091 Final Exam Thursday Feb
More informationCHAPTER 1: Decomposition Methods
CHAPTER 1: Decomposition Methods Prof. Alan Wan 1 / 48 Table of contents 1. Data Types and Causal vs.time Series Models 2 / 48 Types of Data Time series data: a sequence of observations measured over time,
More informationR = µ + Bf Arbitrage Pricing Model, APM
4.2 Arbitrage Pricing Model, APM Empirical evidence indicates that the CAPM beta does not completely explain the cross section of expected asset returns. This suggests that additional factors may be required.
More informationPackage hts. June 21, 2017
Title Hierarchical and Grouped Time Series Version 5.1.4 Package hts June 21, 2017 Methods for analysing and forecasting hierarchical and grouped time series. Depends R (>= 3.0.2), forecast (>= 8.1) Imports
More informationRecent Developments in Benchmarking to Annual Totals in X-12-ARIMA and at Statistics Canada
CONFERENCE ON SEASONALITY, SEASONAL ADJUSTMENT AND THEIR IMPLICATIONS FOR SHORT-TERM ANALYSIS AND FORECASTING 10-12 MAY 2006 Recent Developments in Benchmarking to Annual Totals in X-12-ARIMA and at Statistics
More information