Investigating a new estimator of the serial correlation coefficient

Size: px
Start display at page:

Download "Investigating a new estimator of the serial correlation coefficient"

Transcription

1 Ivestigatig a ew estimator of the serial correlatio coefficiet Mafred Mudelsee Istitute of Mathematics ad Statistics Uiversity of Ket at Caterbury Caterbury Ket CT2 7NF Uited Kigdom 8 May Abstract A ew estimator of the lag-1 serial correlatio coefficiet ρ, ɛ 3 i ɛ i+1 / 1 ɛ 4 i, is compared with the old estimator, 1 ɛ i ɛ i+1 / 1 ɛ 2 i, for a statioary AR(1) process with kow mea. The mea ad the variace of both estimators are calculated usig the secod order Taylor expasio of a ratio. No further approximatio is used. I case of the mea of the old estimator, we derive Marriott ad Pope s (1954) formula, with ( 1) 1 istead of () 1, ad a additioal term ( 1) 2. I case of the variace of the old estimator, Bartlett s (1946) formula results, with ( 1) 1 istead of () 1. The theoretical expressios are corroborated with simulatio experimets. The mai results are as follows. (1) The ew estimator has a larger egative bias ad a larger variace tha the old estimator. (2) The theoretical results for the mea ad the variace of the old estimator describe the pricipal behaviours over the etire rage of ρ, i particular, the declie to zero egative bias as ρ approaches uity.

2 1 Itroductio Cosider the followig estimators of the serial correlatio coefficiet ρ of lag 1 from a time series ɛ i (i = 1,..., ) sampled from a process E i with kow (zero) mea: ˆρ = ˆρ = 1 ɛ i ɛ i+1 1 ɛ 2 i 1 ɛ 3 i ɛ i+1 1 ɛ 4 i, (1). (2) Estimators of type (1) are well-kow (e. g. Bartlett 1946, Marriott ad Pope 1954), whereas (2) is ew to my best kowledge. Let E i be the statioary AR(1) process, E 1 N(0, 1), E i = ρ E i 1 + U i, i = 2,...,, (3) with U i i. i. d. N(0, σ 2 ) ad σ 2 = 1 ρ 2. The followig misjudgemet of mie led to the preset study. ˆρ miimises (ɛ i+1 ρɛ i ) 2 ɛ 2 i, which is a weighted sum of squares. However, sice E i has costat variace uity, o weightig should be ecessary. Nevertheless, we compare estimators (1) ad (2) for process (3). We restrict ourselves to 0 < ρ < 1. We first calculate their variaces, respectively meas, up to the secod order of the Taylor expasio of a ratio, similarly to Bartlett (1946), respectively Marriott ad Pope (1954). Sice we cocetrate o the lag-1 estimators ad process (3), o further approximatio is ecessary. These theoretical expressios ad also those of White (1961) for estimator (1) are the examied with simulatio experimets. 2 Series expasios 2.1 Old estimator Variace We write (1) as ˆρ = ɛ i ɛ i+1 1 ɛ 2 i (4) = c v, say, 1

3 ad have, up to the secod order of deviatios from the meas of c ad v, the variace var(ˆρ) = var(c/v) var(c) E 2 (v) 2 E(c)cov(v, c) E 3 (v) + E 2 (c) var(v) E 4 (v). (5) We ow apply a stadard result for quadravariate stadard Gaussia distributios with serial correlatios ρ j (e. g. Priestley 1981:325), cov(ɛ a ɛ a+s, ɛ b ɛ b+s+t ) = ρ b a ρ b a+t + ρ b a+s+t ρ b a s, from which we derive, for ɛ i followig process (3): cov ( 1 ɛ a ɛ a+s, 1 ) ɛ a=1 b ɛ b+s+t = b=1 = 1 2 (ρ b a ρ b a+t + ρ b a+s+t ρ b a s ). (6) Without further approximatio we ca directly derive the various costituets of the right-had side of (5) by the meas of (6). var(c) = var = 1 2 ( ) 1 ɛ i ɛ i+1 = (put s = 1 ad t = 0 i (6)) ρ 2 b a + ρ b a+1 ρ b a 1. We fid, by summig over the poits of the a-b lattice i a diagoal maer, 2 ρ 2 b a = ( 1) + 2 i ρ 2( i 1) = ( 1) ( ) 2 1 ρ (ρ2 4 ρ 2 ) (7) 2 (1 ρ 2 ) 2 (at the last step we have used the arithmetic-geometric progressio), ad 2 ρ b a+1 ρ b a 1 = ( 1) ρ i ρ 2( i 1) = ( 1) (7) ad (8) give var(c). Further, cov(v, c) = cov ( ) 2 1 ρ + 2 ρ (ρ2 4 ρ 2 ). (8) (1 ρ 2 ) 2 ( 1 2 ɛ 2 i, 1 ɛ i ɛ i+1 )

4 = (put s = 0 ad t = 1 i (6)) = 2 2 ρ b a ρ b a+1. We fid 2 ρ b a ρ b a+1 = ρ (2 i + 1) ρ 2( i) 3 = ( 1) 2 ρ ( ) 1 1 ρ (ρ2 5 ρ 3 ) + (ρ2 3 ρ 1 ). (9) (1 ρ 2 ) 2 (1 ρ 2 ) (At the last step we have used the arithmetic-geometric ad also the geometric progressio.) (9) gives cov(v, c). Further, var(v) = var = 2 2 ( 1 ɛ 2 i ) ρ 2 b a, which is give from (7). We further eed = (put s = t = 0 i (6)) E(v) = 1 (10) ad E(c) = ρ( 1). (11) All three terms cotributig to var(ˆρ) have a part ( 1) 2. However, these parts cacel out: var(ˆρ) (1 ρ2 ) ( 1). (12) Bartlett (1946) already gave, up to the order () 1, which caot be distiguished from our result. var(ˆρ) (1 ρ2 ), (13) 3

5 2.1.2 Mea We have, up to the secod order of deviatios from the meas of c ad v, the mea E(ˆρ) = E(c/v) E(c) cov(v, c) + E(c) var(v) E(v) E 2 (v) E 3 (v). (14) As above, we derive cov(v, c) ad var(v) from (9) ad (7), respectively, ad have E(v) ad E(c) give by (10) ad (11), respectively, yieldig E(ˆρ) ρ 2 ρ ( 1) + 2 (ρ ρ 2 1 ). (15) ( 1) 2 (1 ρ 2 ) Marriott ad Pope (1954) ivestigated the estimator ˆρ = ad gave, up to the order () 1, ɛ 2 i E(ˆρ ) ρ 2 ρ, ɛ i ɛ i+1 which caot be distiguished from the first two terms of our result Ukow mea of the process I have also tried to calculate up to the same order of approximatio the mea ad the variace of the estimator ( 1 ɛi 1 ) ( 1 1 j=1 ɛ j ɛi+1 1 ) 1 1 j=1 ɛ j+1 ( 1 ɛi 1 ) 1 2, 1 j=1 ɛ j for the case whe the mea of the process is ukow. That would require to calculate the followig sums over the poits of a cubic lattice: a,b,c=1 ρ b a ρ c a s for s = 0 ad 1. However, I was uable to obtai exact formulas. 2.2 New estimator Variace We write (2) as ˆρ = ɛ 3 i ɛ i+1 1 ɛ 4 i = d w, say. 4

6 var(ˆρ ), up to the secod order of the Taylor expasio, is give by (5), substitutig d for c ad w for v, respectively. We eed the followig result for octavariate stadard Gaussia distributios with serial correlatios ρ j (Appedix), cov(ɛ 3 aɛ a+s, ɛ 3 bɛ b+s+t ) = 9 ρ b a ρ b a+t + 9 ρ b a+s+t ρ b a s from which we derive, for ɛ i followig process (3): ( 1 cov ɛ 3 aɛ a+s, 1 ) ɛ 3 a=1 bɛ b+s+t = = 1 2 b= ρ s ρ b a ρ b a+s+t + 18 ρ b a ρ b a s ρ s+t + 18 ρ s ρ 2 b a ρ s+t + 18 ρ 2 b a ρ b a+s+t ρ b a s + 6 ρ 3 b a ρ b a+t, (16) ( 9 ρ b a ρ b a+t + 9 ρ b a+s+t ρ b a s + 18 ρ s ρ b a ρ b a+s+t + 18 ρ b a ρ b a s ρ s+t + 18 ρ s ρ 2 b a ρ s+t + 18 ρ 2 b a ρ b a+s+t ρ b a s + 6 ρ 3 b a ρ b a+t ). (17) Without further simplificatio we ca directly obtai the various costituets of var(ˆρ ) by the meas of (17). ( ) 1 var(d) = var ɛ 3 i ɛ i+1 = (put s = 1 ad t = 0 i (17)) = ρ ρ 2 b a ρ ρ 4 b a. ρ b a+1 ρ b a 1 ρ b a ρ b a ρ ρ 2 b a + 18 ρ b a ρ b a 1 ρ 2 b a ρ b a+1 ρ b a 1 The first ad the fifth sum is give by (7), the secod sum by (8) ad the third by (9). We fid, ρ b a ρ b a 1 = 5 ρ b a ρ b a+1,

7 which is give by (9), ad 2 ρ 2 b a ρ b a+1 ρ b a 1 = ( 1) ρ i ρ 4( i 1) = ( 1) ( ) 2 1 ρ + 4 ρ (ρ4 8 ρ 4 ) (18) (1 ρ 4 ) 2 2 ρ 4 b a = ( 1) + 2 i ρ 4( i 1) = ( 1) ( ) 2 1 ρ (ρ4 8 ρ 4 ). (19) 4 (1 ρ 4 ) 2 (7), (8), (9), (18) ad (19) give var(d). Further, cov(w, d) = cov ( 1 ɛ 4 i, 1 ɛ 3 i ɛ i+1 ) = (put s = 0 ad t = 1 i (17)) = The last sum of this expressio, ρ 3 b a ρ b a+1 = ( 1) ρ + (7), (9) ad (20) give cov(w, d). Further, var(w) = var ρ b a ρ b a ρ ρ 3 b a ρ b a+1. 2 i ρ 4( i) 3 + ρ 2 b a 2 ( 1 = ( 1) ρ 1 ρ ρ ρ 1 ) 5 ρ i ρ 4( i) 5 + (ρ4 7 + ρ 4 9 ) (1 ρ 4+4 ) (1 ρ 4 ) 2. (20) ( 1 = ɛ 4 i ) = (put s = t = 0 i (17)) ρ 2 b a ρ 4 b a,

8 which is give from (7) ad (19), respectively. We further eed ad E(w) = E(d) = 3 ( 1) (21) 3 ρ ( 1). (22) As for the old estimator, the terms ( 1) 2 which cotribute to var(ˆρ ) cacel out: Mea var(ˆρ ) 5 (1 ρ 2 ) 3 ( 1). (23) We use (14) with d istead of c ad w istead of v, respectively. We derive cov(w, d) from (7), (9) ad (20), further, var(w) from (7) ad (19), ad we have E(w) ad E(d) give by (21) ad (22), respectively. This yields, up to the secod order of deviatios from the meas of d ad w, ( E(ˆρ 1 4 ) ρ ( 1) 3 ρ 5 2 ) 1 + ρ Remark + [ 1 4 (ρ ρ2 1 ) ( 1) 2 (1 ρ 2 ) (ρ 3 ρ 4 1 ] ). (24) (1 ρ 2 ) (1 + ρ 2 ) 2 ˆρ is foud to have a larger egative bias ad also a larger variace tha ˆρ. I the simulatio experimet, we ited ot oly to prove that. We are also iterested to examie the sigificace of the term ( 1) 2 i (15). I the compariso we iclude the theoretical results of White (1961) who studied the old estimator (1) for process (3). He calculated the kth momet of (c/v) via the joit momet geeratig fuctio m(c, v), with c ad v defied as i (4) without the factors 1/, E(ˆρ k ) = 0 vk v2... k m(c, v) c k dv 1 dv 2... dv k. c = 0 He expaded the itegrad to terms of order 3 ad ρ 4 ad gave the followig results (the idex W refers to his study): var(ˆρ) W ( ) ( ) ρ ρ4, (25) E(ˆρ) W ( ) ρ ρ ρ5. (26) 7

9 3 Simulatio experimets I the first experimet, for each combiatio of ad ρ listed i Table 1, we geerated a set of time series from process (3). For every time series, ˆρ ad ˆρ have bee calculated. The sample meas, µˆρ sim ad µˆρ sim, ad the sample stadard deviatios, σˆρ sim ad σˆρ sim, over the simulatios are listed i Table 1. Therei, our theoretical values from (12), (15), (23) ad (24) are icluded. I case of the meas, these are additioally splitted ito terms ( 1) 2 ad terms of lower order. Also the theoretical values from White s (1961) formula, herei (25) ad (26), are listed. We further ivestigated whether the distributio of ˆρ approaches Gaussiaity as fast as that of ˆρ. Fig. 1 shows histograms of ˆρ, respectively ˆρ, from the first simulatio experimet i compariso with Gaussia distributios N(µˆρ sim, σ 2ˆρ sim), respectively N(µˆρ sim, σ 2ˆρ sim). I the secod simulatio experimet (Fig. 2), formulas (12) versus (25) ad (15) versus (26) are examied over the etire rage of ρ, with particular iterest o ρ large. We plot the egative bias, ρ E(ˆρ), ad var(ˆρ). For each combiatio of ρ ad, the umber of simulated time series after (3) is The error due to the limited umber of simulatios is small eough for ot ifluecig ay iteded compariso. 4 Results ad discussio The first simulatio experimet (Table 1) cofirms, over a broad rage of ρ ad, that ˆρ has a larger bias ad a larger variace, respectively, tha ˆρ. I geeral, the deviatios betwee the theoretical results ad the simulatio results decrease with icreased, as is to be expected. For ay combiatio of ρ ad listed i Table 1, our terms ( 1) 2 i (15), respectively (24), brig these theoretical results closer to the simulatio results, with the exceptio of E(ˆρ) for ρ = 0.9 ad = 800 where the simulatio oise prevets that compariso. For ρ large, respectively small, these secod order terms cotribute heavier. The frequecy distributio of ˆρ (Fig. 1) has a similar shape as that of ˆρ. It is shifted to smaller values agaist that ad also broader, reflectig the larger egative bias, respectively the larger variace. The fuctioal form of the distributio of ˆρ is approximately the Leipik distributio, which is heavier skewed for ρ large ad teds to Gaussiaity with icreased. Both estimators seem to approach Gaussiaity equally fast (Fig. 1). The secod simulatio experimet (Fig. 2) compares White s (1961) theoretical results for ˆρ, (25) ad (26), with ours, (12) ad (15). It shows that his are better performig up to a certai value of ρ. Above, our formulas describe better the simulatio, particularly, the de- 8

10 clie to zero egative bias, respectively zero variace, as ρ approaches uity. The declie to zero egative bias is caused by the term ( 1) 2 i (15). Those declies are reasoable, sice for ρ = 1 all time series poits have equal value ad c = v i (4). For small ρ, his formulas perform better sice they are more accurate with respect to powers of (1/) tha ours. For larger ρ his approximatio becomes less accurate. I particular, for 3 ad ρ > 0, ρ E(ˆρ) W caot d become zero. For 10, (ρ E(ˆρ) dρ W ) caot become egative. For 8, var(ˆρ) W caot become zero. That meas, for those cases White s (1961) formulas caot produce the declie to zero egative bias ad zero variace, respectively. The fact that Bartlett s (1946) formula for the variace, (13), describes the simulatio better tha (12) for ρ 0, is regarded as spurious. These results mea that the expasio formulas for var(ˆρ) ad E(ˆρ), (5) ad (14), respectively, are sufficiet to derive the pricipal behaviours for ρ 1. I case of the mea, however, o further approximatio is allowed which would lead to the term ( 1) 2 i (15) be eglected. 5 Coclusios 1. I case of the statioary AR(1) process (3) with kow mea, the ew estimator, ˆρ, has a larger egative bias ad a larger variace tha the old estimator, ˆρ. Its distributio teds to Gaussiaity about equally fast. 2. Our formula (15) for E(ˆρ) describes the expected declie to zero egative bias for ρ The formula (12) for var(ˆρ) describes the expected declie to zero variace for ρ The secod order Taylor expasio of a ratio is sufficiet to derive these two pricipal behaviours for ρ 1, if o further approximatios are made. This coditio could be fulfilled sice we had cocetrated ourselves o the lag-1 estimator ad process (3). 5. For ukow mea of the process, it is more complicated to derive the equatios with the same accuracy. Ackowledgemets Q. Yao is appreciated for commets o the mauscript. The preset study was carried out while the author was Marie Curie Research Fellow (EU-TMR grat ERBFMBICT971919). 9

11 Refereces Bartlett, M. S., 1946, O the theoretical specificatio ad samplig properties of autocorrelated time-series: Joural of the Royal Statistical Society Supplemet, v. 8, p (Corrigeda: 1948, Joural of the Royal Statistical Society Series B, v. 10, o. 1). Marriott, F. H. C., ad Pope, J. A., 1954, Bias i the estimatio of autocorrelatios: Biometrika, v. 41, p Priestley, M. B., 1981, Spectral Aalysis ad Time Series: Lodo, Academic Press, 890 p. Scott, D. W., 1979, O optimal ad data-based histograms: Biometrika, v. 66, p White, J. S., 1961, Asymptotic expasios for the mea ad variace of the serial correlatio coefficiet: Biometrika, v. 48, p Appedix We assume that ɛ i is draw from a stadard Gaussia distributio with serial correlatios ρ j. We derive (16) by meas of the momet geeratig fuctio. Write cov(ɛ 3 aɛ a+s, ɛ 3 bɛ b+s+t ) = E(ɛ 3 aɛ a+s ɛ 3 bɛ b+s+t ) E(ɛ 3 aɛ a+s ) E(ɛ 3 bɛ b+s+t ) = E(ɛ 3 aɛ a+s ɛ 3 bɛ b+s+t ) 9 ρ s ρ s+t. Now, E(ɛ 3 aɛ a+s ɛ 3 bɛ b+s+t ) is the coefficiet of (t 3 1 t 2 t 3 3 t 4 )/(3! 1! 3! 1!) i the momet geeratig fuctio with m(t 1, t 2, t 3, t 4 ) = exp σ 2 ij t i t j, j=1 σ 11 = σ 22 = σ 33 = σ 44 = 1, σ 12 = σ 21 = ρ s, σ 13 = σ 31 = ρ b a, σ 14 = σ 41 = ρ b a+s+t, σ 23 = σ 32 = ρ b a s, 10

12 σ 24 = σ 42 = ρ b a+t, σ 34 = σ 43 = ρ s+t. Terms (t 3 1 t 2 t 3 3 t 4 ) i m(t 1, t 2, t 3, t 4 ) ca oly be withi the fourth order of the series expasio of the expoetial fuctio, i. e., withi further restrictig, withi σ 4! 2 ij t i t j, j= (t2 1 + t σ 12 t 1 t σ 13 t 1 t 3 further restrictig, withi + 2 σ 14 t 1 t σ 23 t 2 t σ 24 t 2 t σ 34 t 3 t 4 ) 4, (4 σ2 13t 2 1t t 2 1t σ 12 t 3 1t σ 13 t 3 1t σ 14 t 3 1t σ 23 t 2 1t 2 t σ 24 t 2 1t 2 t σ 34 t 2 1t 3 t σ 12 t 1 t 2 t σ 13 t 1 t σ 14 t 1 t 2 3t σ 23 t 2 t σ 24 t 2 t 2 3t σ 34 t 3 3t σ 12 σ 13 t 2 1t 2 t σ 12 σ 14 t 2 1t 2 t σ 12 σ 34 t 1 t 2 t 3 t σ 13 σ 14 t 2 1t 3 t σ 13 σ 23 t 1 t 2 t σ 13 σ 24 t 1 t 2 t 3 t σ 13 σ 34 t 1 t 2 3t σ 14 σ 23 t 1 t 2 t 3 t σ 23 σ 34 t 2 t 2 3t 4 ) 2. Fially, these terms are (2 4 σ2 13t 2 1t σ 12 σ 34 t 1 t 2 t 3 t σ 2 13t 2 1t σ 13 σ 24 t 1 t 2 t 3 t σ 2 13t 2 1t σ 14 σ 23 t 1 t 2 t 3 t t 2 1t σ 12 σ 34 t 1 t 2 t 3 t t 2 1t σ 13 σ 24 t 1 t 2 t 3 t t 2 1t σ 14 σ 23 t 1 t 2 t 3 t σ 12 t 3 1t 2 4 σ 34 t 3 3t σ 13 t 3 1t 3 4 σ 24 t 2 t 2 3t σ 13 t 3 1t 3 8 σ 23 σ 34 t 2 t 2 3t σ 14 t 3 1t 4 4 σ 23 t 2 t σ 23 t 2 1t 2 t 3 4 σ 14 t 1 t 2 3t σ 23 t 2 1t 2 t 3 8 σ 13 σ 34 t 1 t 2 3t σ 24 t 2 1t 2 t 4 4 σ 13 t 1 t σ 34 t 2 1t 3 t 4 4 σ 12 t 1 t 2 t σ 34 t 2 1t 3 t 4 8 σ 13 σ 23 t 1 t 2 t σ 12 t 1 t 2 t σ 13 σ 14 t 2 1t 3 t σ 13 t 1 t σ 12 σ 14 t 2 1t 2 t σ 14 t 1 t 2 3t 4 8 σ 12 σ 13 t 2 1t 2 t σ 12 σ 13 t 2 1t 2 t 3 8 σ 13 σ 34 t 1 t 2 3t σ 13 σ 14 t 2 1t 3 t 4 8 σ 13 σ 23 t 1 t 2 t 2 3). 11

13 Thus, we fid E(ɛ 3 aɛ a+s ɛ 3 bɛ b+s+t ) = 3! 1! 3! 1! (96 σ σ σ 13 σ σ 14 σ σ 12 σ 13 σ σ 13 σ 23 σ σ 12 σ 2 13σ σ 2 13σ 2 14σ σ 3 13σ 24 ). This gives the fial result cov(ɛ 3 aɛ a+s, ɛ 3 bɛ b+s+t ) = 9 ρ b a ρ b a+t + 9 ρ b a+s+t ρ b a s + 18 ρ s ρ b a ρ b a+s+t + 18 ρ b a ρ b a s ρ s+t + 18 ρ s ρ 2 b a ρ s+t + 18 ρ 2 b a ρ b a+s+t ρ b a s + 6 ρ 3 b a ρ b a+t. It should be oted that this result has bee checked usig the joit cumulat of order eight, i my assessmet without less effort. 12

14 Table 1: First simulatio experimet. The umber of simulated time series after (3) is i each case S.d.: stadard deviatio. The order of the ucertaity of the mea, respectively the stadard deviatio, due to the limited umber of simulatios, is S.d. / Sim: simulatio (sample mea ad sample stadard deviatio, µˆρ sim ad σˆρ sim, respectively µˆρ sim ad σˆρ sim ). T(24, 23): theoretical value, this study, (24), respectively (23). 2d: term ( 1) 2 i (15), respectively (24). 1st: terms ot ( 1) 2 i (15), respectively (24). ˆρ Sim ˆρ Sim ˆρ T(24) 1st ˆρ T(24) 2d ˆρ T(24, 23) ˆρ T(15) 1st ˆρ T(15) 2d ˆρ T(15, 12) ˆρ T(26, 25) ρ =.200 Mea = 10 S.d S.d. / ρ =.200 Mea = 20 S.d S.d. / ρ =.200 Mea = 50 S.d S.d. / ρ =.500 Mea = 10 S.d S.d. / ρ =.500 Mea = 50 S.d S.d. / ρ =.500 Mea = 150 S.d S.d. /

15 Table 1: (Cotiued) ˆρ Sim ˆρ Sim ˆρ T(24) 1st ˆρ T(24) 2d ˆρ T(24, 23) ˆρ T(15) 1st ˆρ T(15) 2d ˆρ T(15, 12) ˆρ T(26, 25) ρ =.900 Mea = 20 S.d S.d. / ρ =.900 Mea = 100 S.d S.d. / ρ =.900 Mea = 800 S.d S.d. / ρ =.980 Mea = 20 S.d S.d. / ρ =.980 Mea = 200 S.d S.d. / ρ =.980 Mea = 2000 S.d S.d. / ρ =.999 Mea = 500 S.d S.d. / ρ =.999 Mea = 2000 S.d S.d. / ρ =.999 Mea = S.d S.d. /

16 ρ = 0.20 = ρ = 0.20 = ρ = 0.20 = Figure 1: First simulatio experimet (cf. Table 1). Histograms of ˆρ (thick lie), respectively ˆρ (thi lie), compared with Gaussia distributios N(µˆρ sim, σ 2ˆρ sim) (heavy lie), respectively N(µˆρ sim, σ 2ˆρ sim) (light lie). The umber of histogram classes follows Scott (1979). 15

17 ρ = 0.50 = ρ = 0.50 = ρ = 0.50 = Figure 1: (Cotiued) 16

18 ρ = 0.90 = ρ = 0.90 = ρ = 0.90 = 800 Figure 1: (Cotiued)

19 ρ = 0.98 = ρ = 0.98 = ρ = 0.98 = Figure 1: (Cotiued) 18

20 ρ = = ρ = = ρ = = Figure 1: (Cotiued)

21 ρ - E(ρ^) 0.10 = SQRT[var(ρ^)] = ρ Figure 2: Secod simulatio experimet. Above: egative bias, below: stadard deviatio. Simulatio results (dots). Theoretical result, our formula, (12) ad (15), respectively, (heavy lie). Theoretical result, White s (1961) formula, (25) ad (26), respectively, (light lie). 20

22 ρ - E(ρ^) = SQRT[var(ρ^)] 0.00 = ρ Figure 2: (Cotiued) 21

23 ρ - E(ρ^) = SQRT[var(ρ^)] 0.00 = ρ Figure 2: (Cotiued) 22

Topic 9: Sampling Distributions of Estimators

Topic 9: Sampling Distributions of Estimators Topic 9: Samplig Distributios of Estimators Course 003, 2016 Page 0 Samplig distributios of estimators Sice our estimators are statistics (particular fuctios of radom variables), their distributio ca be

More information

Statistical Analysis on Uncertainty for Autocorrelated Measurements and its Applications to Key Comparisons

Statistical Analysis on Uncertainty for Autocorrelated Measurements and its Applications to Key Comparisons Statistical Aalysis o Ucertaity for Autocorrelated Measuremets ad its Applicatios to Key Comparisos Nie Fa Zhag Natioal Istitute of Stadards ad Techology Gaithersburg, MD 0899, USA Outlies. Itroductio.

More information

CEE 522 Autumn Uncertainty Concepts for Geotechnical Engineering

CEE 522 Autumn Uncertainty Concepts for Geotechnical Engineering CEE 5 Autum 005 Ucertaity Cocepts for Geotechical Egieerig Basic Termiology Set A set is a collectio of (mutually exclusive) objects or evets. The sample space is the (collectively exhaustive) collectio

More information

A goodness-of-fit test based on the empirical characteristic function and a comparison of tests for normality

A goodness-of-fit test based on the empirical characteristic function and a comparison of tests for normality A goodess-of-fit test based o the empirical characteristic fuctio ad a compariso of tests for ormality J. Marti va Zyl Departmet of Mathematical Statistics ad Actuarial Sciece, Uiversity of the Free State,

More information

Lecture 6 Chi Square Distribution (χ 2 ) and Least Squares Fitting

Lecture 6 Chi Square Distribution (χ 2 ) and Least Squares Fitting Lecture 6 Chi Square Distributio (χ ) ad Least Squares Fittig Chi Square Distributio (χ ) Suppose: We have a set of measuremets {x 1, x, x }. We kow the true value of each x i (x t1, x t, x t ). We would

More information

Topic 9: Sampling Distributions of Estimators

Topic 9: Sampling Distributions of Estimators Topic 9: Samplig Distributios of Estimators Course 003, 2018 Page 0 Samplig distributios of estimators Sice our estimators are statistics (particular fuctios of radom variables), their distributio ca be

More information

Lecture 6 Chi Square Distribution (χ 2 ) and Least Squares Fitting

Lecture 6 Chi Square Distribution (χ 2 ) and Least Squares Fitting Lecture 6 Chi Square Distributio (χ ) ad Least Squares Fittig Chi Square Distributio (χ ) Suppose: We have a set of measuremets {x 1, x, x }. We kow the true value of each x i (x t1, x t, x t ). We would

More information

Since X n /n P p, we know that X n (n. Xn (n X n ) Using the asymptotic result above to obtain an approximation for fixed n, we obtain

Since X n /n P p, we know that X n (n. Xn (n X n ) Using the asymptotic result above to obtain an approximation for fixed n, we obtain Assigmet 9 Exercise 5.5 Let X biomial, p, where p 0, 1 is ukow. Obtai cofidece itervals for p i two differet ways: a Sice X / p d N0, p1 p], the variace of the limitig distributio depeds oly o p. Use the

More information

Topic 9: Sampling Distributions of Estimators

Topic 9: Sampling Distributions of Estimators Topic 9: Samplig Distributios of Estimators Course 003, 2018 Page 0 Samplig distributios of estimators Sice our estimators are statistics (particular fuctios of radom variables), their distributio ca be

More information

A statistical method to determine sample size to estimate characteristic value of soil parameters

A statistical method to determine sample size to estimate characteristic value of soil parameters A statistical method to determie sample size to estimate characteristic value of soil parameters Y. Hojo, B. Setiawa 2 ad M. Suzuki 3 Abstract Sample size is a importat factor to be cosidered i determiig

More information

Estimation of the Mean and the ACVF

Estimation of the Mean and the ACVF Chapter 5 Estimatio of the Mea ad the ACVF A statioary process {X t } is characterized by its mea ad its autocovariace fuctio γ ), ad so by the autocorrelatio fuctio ρ ) I this chapter we preset the estimators

More information

Random Variables, Sampling and Estimation

Random Variables, Sampling and Estimation Chapter 1 Radom Variables, Samplig ad Estimatio 1.1 Itroductio This chapter will cover the most importat basic statistical theory you eed i order to uderstad the ecoometric material that will be comig

More information

4. Partial Sums and the Central Limit Theorem

4. Partial Sums and the Central Limit Theorem 1 of 10 7/16/2009 6:05 AM Virtual Laboratories > 6. Radom Samples > 1 2 3 4 5 6 7 4. Partial Sums ad the Cetral Limit Theorem The cetral limit theorem ad the law of large umbers are the two fudametal theorems

More information

Investigating the Significance of a Correlation Coefficient using Jackknife Estimates

Investigating the Significance of a Correlation Coefficient using Jackknife Estimates Iteratioal Joural of Scieces: Basic ad Applied Research (IJSBAR) ISSN 2307-4531 (Prit & Olie) http://gssrr.org/idex.php?joural=jouralofbasicadapplied ---------------------------------------------------------------------------------------------------------------------------

More information

Chapter 10: Power Series

Chapter 10: Power Series Chapter : Power Series 57 Chapter Overview: Power Series The reaso series are part of a Calculus course is that there are fuctios which caot be itegrated. All power series, though, ca be itegrated because

More information

Expectation and Variance of a random variable

Expectation and Variance of a random variable Chapter 11 Expectatio ad Variace of a radom variable The aim of this lecture is to defie ad itroduce mathematical Expectatio ad variace of a fuctio of discrete & cotiuous radom variables ad the distributio

More information

Chapter 6 Sampling Distributions

Chapter 6 Sampling Distributions Chapter 6 Samplig Distributios 1 I most experimets, we have more tha oe measuremet for ay give variable, each measuremet beig associated with oe radomly selected a member of a populatio. Hece we eed to

More information

1 Inferential Methods for Correlation and Regression Analysis

1 Inferential Methods for Correlation and Regression Analysis 1 Iferetial Methods for Correlatio ad Regressio Aalysis I the chapter o Correlatio ad Regressio Aalysis tools for describig bivariate cotiuous data were itroduced. The sample Pearso Correlatio Coefficiet

More information

Homework 3 Solutions

Homework 3 Solutions Math 4506 Sprig 04 Homework 3 Solutios. a The ACF of a MA process has a o-zero value oly at lags, 0, ad. Problem 4.3 from the textbook which you did t do, so I did t expect you to metio this shows that

More information

Approximate Confidence Interval for the Reciprocal of a Normal Mean with a Known Coefficient of Variation

Approximate Confidence Interval for the Reciprocal of a Normal Mean with a Known Coefficient of Variation Metodološki zvezki, Vol. 13, No., 016, 117-130 Approximate Cofidece Iterval for the Reciprocal of a Normal Mea with a Kow Coefficiet of Variatio Wararit Paichkitkosolkul 1 Abstract A approximate cofidece

More information

SNAP Centre Workshop. Basic Algebraic Manipulation

SNAP Centre Workshop. Basic Algebraic Manipulation SNAP Cetre Workshop Basic Algebraic Maipulatio 8 Simplifyig Algebraic Expressios Whe a expressio is writte i the most compact maer possible, it is cosidered to be simplified. Not Simplified: x(x + 4x)

More information

TR/46 OCTOBER THE ZEROS OF PARTIAL SUMS OF A MACLAURIN EXPANSION A. TALBOT

TR/46 OCTOBER THE ZEROS OF PARTIAL SUMS OF A MACLAURIN EXPANSION A. TALBOT TR/46 OCTOBER 974 THE ZEROS OF PARTIAL SUMS OF A MACLAURIN EXPANSION by A. TALBOT .. Itroductio. A problem i approximatio theory o which I have recetly worked [] required for its solutio a proof that the

More information

Statistical Inference (Chapter 10) Statistical inference = learn about a population based on the information provided by a sample.

Statistical Inference (Chapter 10) Statistical inference = learn about a population based on the information provided by a sample. Statistical Iferece (Chapter 10) Statistical iferece = lear about a populatio based o the iformatio provided by a sample. Populatio: The set of all values of a radom variable X of iterest. Characterized

More information

ARIMA Models. Dan Saunders. y t = φy t 1 + ɛ t

ARIMA Models. Dan Saunders. y t = φy t 1 + ɛ t ARIMA Models Da Sauders I will discuss models with a depedet variable y t, a potetially edogeous error term ɛ t, ad a exogeous error term η t, each with a subscript t deotig time. With just these three

More information

Summary: CORRELATION & LINEAR REGRESSION. GC. Students are advised to refer to lecture notes for the GC operations to obtain scatter diagram.

Summary: CORRELATION & LINEAR REGRESSION. GC. Students are advised to refer to lecture notes for the GC operations to obtain scatter diagram. Key Cocepts: 1) Sketchig of scatter diagram The scatter diagram of bivariate (i.e. cotaiig two variables) data ca be easily obtaied usig GC. Studets are advised to refer to lecture otes for the GC operatios

More information

ECE 901 Lecture 12: Complexity Regularization and the Squared Loss

ECE 901 Lecture 12: Complexity Regularization and the Squared Loss ECE 90 Lecture : Complexity Regularizatio ad the Squared Loss R. Nowak 5/7/009 I the previous lectures we made use of the Cheroff/Hoeffdig bouds for our aalysis of classifier errors. Hoeffdig s iequality

More information

Goodness-Of-Fit For The Generalized Exponential Distribution. Abstract

Goodness-Of-Fit For The Generalized Exponential Distribution. Abstract Goodess-Of-Fit For The Geeralized Expoetial Distributio By Amal S. Hassa stitute of Statistical Studies & Research Cairo Uiversity Abstract Recetly a ew distributio called geeralized expoetial or expoetiated

More information

ACCESS TO SCIENCE, ENGINEERING AND AGRICULTURE: MATHEMATICS 1 MATH00030 SEMESTER / Statistics

ACCESS TO SCIENCE, ENGINEERING AND AGRICULTURE: MATHEMATICS 1 MATH00030 SEMESTER / Statistics ACCESS TO SCIENCE, ENGINEERING AND AGRICULTURE: MATHEMATICS 1 MATH00030 SEMESTER 1 018/019 DR. ANTHONY BROWN 8. Statistics 8.1. Measures of Cetre: Mea, Media ad Mode. If we have a series of umbers the

More information

1 Introduction to reducing variance in Monte Carlo simulations

1 Introduction to reducing variance in Monte Carlo simulations Copyright c 010 by Karl Sigma 1 Itroductio to reducig variace i Mote Carlo simulatios 11 Review of cofidece itervals for estimatig a mea I statistics, we estimate a ukow mea µ = E(X) of a distributio by

More information

Lecture 3. Properties of Summary Statistics: Sampling Distribution

Lecture 3. Properties of Summary Statistics: Sampling Distribution Lecture 3 Properties of Summary Statistics: Samplig Distributio Mai Theme How ca we use math to justify that our umerical summaries from the sample are good summaries of the populatio? Lecture Summary

More information

11 Correlation and Regression

11 Correlation and Regression 11 Correlatio Regressio 11.1 Multivariate Data Ofte we look at data where several variables are recorded for the same idividuals or samplig uits. For example, at a coastal weather statio, we might record

More information

Estimation for Complete Data

Estimation for Complete Data Estimatio for Complete Data complete data: there is o loss of iformatio durig study. complete idividual complete data= grouped data A complete idividual data is the oe i which the complete iformatio of

More information

Q-BINOMIALS AND THE GREATEST COMMON DIVISOR. Keith R. Slavin 8474 SW Chevy Place, Beaverton, Oregon 97008, USA.

Q-BINOMIALS AND THE GREATEST COMMON DIVISOR. Keith R. Slavin 8474 SW Chevy Place, Beaverton, Oregon 97008, USA. INTEGERS: ELECTRONIC JOURNAL OF COMBINATORIAL NUMBER THEORY 8 2008, #A05 Q-BINOMIALS AND THE GREATEST COMMON DIVISOR Keith R. Slavi 8474 SW Chevy Place, Beaverto, Orego 97008, USA slavi@dsl-oly.et Received:

More information

Estimation of Backward Perturbation Bounds For Linear Least Squares Problem

Estimation of Backward Perturbation Bounds For Linear Least Squares Problem dvaced Sciece ad Techology Letters Vol.53 (ITS 4), pp.47-476 http://dx.doi.org/.457/astl.4.53.96 Estimatio of Bacward Perturbatio Bouds For Liear Least Squares Problem Xixiu Li School of Natural Scieces,

More information

1 6 = 1 6 = + Factorials and Euler s Gamma function

1 6 = 1 6 = + Factorials and Euler s Gamma function Royal Holloway Uiversity of Lodo Departmet of Physics Factorials ad Euler s Gamma fuctio Itroductio The is a self-cotaied part of the course dealig, essetially, with the factorial fuctio ad its geeralizatio

More information

Properties and Hypothesis Testing

Properties and Hypothesis Testing Chapter 3 Properties ad Hypothesis Testig 3.1 Types of data The regressio techiques developed i previous chapters ca be applied to three differet kids of data. 1. Cross-sectioal data. 2. Time series data.

More information

On a Smarandache problem concerning the prime gaps

On a Smarandache problem concerning the prime gaps O a Smaradache problem cocerig the prime gaps Felice Russo Via A. Ifate 7 6705 Avezzao (Aq) Italy felice.russo@katamail.com Abstract I this paper, a problem posed i [] by Smaradache cocerig the prime gaps

More information

Final Examination Solutions 17/6/2010

Final Examination Solutions 17/6/2010 The Islamic Uiversity of Gaza Faculty of Commerce epartmet of Ecoomics ad Political Scieces A Itroductio to Statistics Course (ECOE 30) Sprig Semester 009-00 Fial Eamiatio Solutios 7/6/00 Name: I: Istructor:

More information

EECS564 Estimation, Filtering, and Detection Hwk 2 Solns. Winter p θ (z) = (2θz + 1 θ), 0 z 1

EECS564 Estimation, Filtering, and Detection Hwk 2 Solns. Winter p θ (z) = (2θz + 1 θ), 0 z 1 EECS564 Estimatio, Filterig, ad Detectio Hwk 2 Sols. Witer 25 4. Let Z be a sigle observatio havig desity fuctio where. p (z) = (2z + ), z (a) Assumig that is a oradom parameter, fid ad plot the maximum

More information

Exercise 4.3 Use the Continuity Theorem to prove the Cramér-Wold Theorem, Theorem. (1) φ a X(1).

Exercise 4.3 Use the Continuity Theorem to prove the Cramér-Wold Theorem, Theorem. (1) φ a X(1). Assigmet 7 Exercise 4.3 Use the Cotiuity Theorem to prove the Cramér-Wold Theorem, Theorem 4.12. Hit: a X d a X implies that φ a X (1) φ a X(1). Sketch of solutio: As we poited out i class, the oly tricky

More information

Comparison Study of Series Approximation. and Convergence between Chebyshev. and Legendre Series

Comparison Study of Series Approximation. and Convergence between Chebyshev. and Legendre Series Applied Mathematical Scieces, Vol. 7, 03, o. 6, 3-337 HIKARI Ltd, www.m-hikari.com http://d.doi.org/0.988/ams.03.3430 Compariso Study of Series Approimatio ad Covergece betwee Chebyshev ad Legedre Series

More information

Confidence Interval for Standard Deviation of Normal Distribution with Known Coefficients of Variation

Confidence Interval for Standard Deviation of Normal Distribution with Known Coefficients of Variation Cofidece Iterval for tadard Deviatio of Normal Distributio with Kow Coefficiets of Variatio uparat Niwitpog Departmet of Applied tatistics, Faculty of Applied ciece Kig Mogkut s Uiversity of Techology

More information

6.3 Testing Series With Positive Terms

6.3 Testing Series With Positive Terms 6.3. TESTING SERIES WITH POSITIVE TERMS 307 6.3 Testig Series With Positive Terms 6.3. Review of what is kow up to ow I theory, testig a series a i for covergece amouts to fidig the i= sequece of partial

More information

The z-transform. 7.1 Introduction. 7.2 The z-transform Derivation of the z-transform: x[n] = z n LTI system, h[n] z = re j

The z-transform. 7.1 Introduction. 7.2 The z-transform Derivation of the z-transform: x[n] = z n LTI system, h[n] z = re j The -Trasform 7. Itroductio Geeralie the complex siusoidal represetatio offered by DTFT to a represetatio of complex expoetial sigals. Obtai more geeral characteristics for discrete-time LTI systems. 7.

More information

THE KALMAN FILTER RAUL ROJAS

THE KALMAN FILTER RAUL ROJAS THE KALMAN FILTER RAUL ROJAS Abstract. This paper provides a getle itroductio to the Kalma filter, a umerical method that ca be used for sesor fusio or for calculatio of trajectories. First, we cosider

More information

PH 425 Quantum Measurement and Spin Winter SPINS Lab 1

PH 425 Quantum Measurement and Spin Winter SPINS Lab 1 PH 425 Quatum Measuremet ad Spi Witer 23 SPIS Lab Measure the spi projectio S z alog the z-axis This is the experimet that is ready to go whe you start the program, as show below Each atom is measured

More information

Module 1 Fundamentals in statistics

Module 1 Fundamentals in statistics Normal Distributio Repeated observatios that differ because of experimetal error ofte vary about some cetral value i a roughly symmetrical distributio i which small deviatios occur much more frequetly

More information

Statistical Fundamentals and Control Charts

Statistical Fundamentals and Control Charts Statistical Fudametals ad Cotrol Charts 1. Statistical Process Cotrol Basics Chace causes of variatio uavoidable causes of variatios Assigable causes of variatio large variatios related to machies, materials,

More information

1 of 7 7/16/2009 6:06 AM Virtual Laboratories > 6. Radom Samples > 1 2 3 4 5 6 7 6. Order Statistics Defiitios Suppose agai that we have a basic radom experimet, ad that X is a real-valued radom variable

More information

Discrete Orthogonal Moment Features Using Chebyshev Polynomials

Discrete Orthogonal Moment Features Using Chebyshev Polynomials Discrete Orthogoal Momet Features Usig Chebyshev Polyomials R. Mukuda, 1 S.H.Og ad P.A. Lee 3 1 Faculty of Iformatio Sciece ad Techology, Multimedia Uiversity 75450 Malacca, Malaysia. Istitute of Mathematical

More information

Improved Class of Ratio -Cum- Product Estimators of Finite Population Mean in two Phase Sampling

Improved Class of Ratio -Cum- Product Estimators of Finite Population Mean in two Phase Sampling Global Joural of Sciece Frotier Research: F Mathematics ad Decisio Scieces Volume 4 Issue 2 Versio.0 Year 204 Type : Double Blid Peer Reviewed Iteratioal Research Joural Publisher: Global Jourals Ic. (USA

More information

MOST PEOPLE WOULD RATHER LIVE WITH A PROBLEM THEY CAN'T SOLVE, THAN ACCEPT A SOLUTION THEY CAN'T UNDERSTAND.

MOST PEOPLE WOULD RATHER LIVE WITH A PROBLEM THEY CAN'T SOLVE, THAN ACCEPT A SOLUTION THEY CAN'T UNDERSTAND. XI-1 (1074) MOST PEOPLE WOULD RATHER LIVE WITH A PROBLEM THEY CAN'T SOLVE, THAN ACCEPT A SOLUTION THEY CAN'T UNDERSTAND. R. E. D. WOOLSEY AND H. S. SWANSON XI-2 (1075) STATISTICAL DECISION MAKING Advaced

More information

ENGI Series Page 6-01

ENGI Series Page 6-01 ENGI 3425 6 Series Page 6-01 6. Series Cotets: 6.01 Sequeces; geeral term, limits, covergece 6.02 Series; summatio otatio, covergece, divergece test 6.03 Stadard Series; telescopig series, geometric series,

More information

CS434a/541a: Pattern Recognition Prof. Olga Veksler. Lecture 5

CS434a/541a: Pattern Recognition Prof. Olga Veksler. Lecture 5 CS434a/54a: Patter Recogitio Prof. Olga Veksler Lecture 5 Today Itroductio to parameter estimatio Two methods for parameter estimatio Maimum Likelihood Estimatio Bayesia Estimatio Itroducto Bayesia Decisio

More information

Resampling Methods. X (1/2), i.e., Pr (X i m) = 1/2. We order the data: X (1) X (2) X (n). Define the sample median: ( n.

Resampling Methods. X (1/2), i.e., Pr (X i m) = 1/2. We order the data: X (1) X (2) X (n). Define the sample median: ( n. Jauary 1, 2019 Resamplig Methods Motivatio We have so may estimators with the property θ θ d N 0, σ 2 We ca also write θ a N θ, σ 2 /, where a meas approximately distributed as Oce we have a cosistet estimator

More information

Chapter 10 Advanced Topics in Random Processes

Chapter 10 Advanced Topics in Random Processes ery Stark ad Joh W. Woods, Probability, Statistics, ad Radom Variables for Egieers, 4th ed., Pearso Educatio Ic.,. ISBN 978--3-33-6 Chapter Advaced opics i Radom Processes Sectios. Mea-Square (m.s.) Calculus

More information

A NEW CLASS OF 2-STEP RATIONAL MULTISTEP METHODS

A NEW CLASS OF 2-STEP RATIONAL MULTISTEP METHODS Jural Karya Asli Loreka Ahli Matematik Vol. No. (010) page 6-9. Jural Karya Asli Loreka Ahli Matematik A NEW CLASS OF -STEP RATIONAL MULTISTEP METHODS 1 Nazeeruddi Yaacob Teh Yua Yig Norma Alias 1 Departmet

More information

Lesson 10: Limits and Continuity

Lesson 10: Limits and Continuity www.scimsacademy.com Lesso 10: Limits ad Cotiuity SCIMS Academy 1 Limit of a fuctio The cocept of limit of a fuctio is cetral to all other cocepts i calculus (like cotiuity, derivative, defiite itegrals

More information

Statisticians use the word population to refer the total number of (potential) observations under consideration

Statisticians use the word population to refer the total number of (potential) observations under consideration 6 Samplig Distributios Statisticias use the word populatio to refer the total umber of (potetial) observatios uder cosideratio The populatio is just the set of all possible outcomes i our sample space

More information

Chapter 2 The Monte Carlo Method

Chapter 2 The Monte Carlo Method Chapter 2 The Mote Carlo Method The Mote Carlo Method stads for a broad class of computatioal algorithms that rely o radom sampligs. It is ofte used i physical ad mathematical problems ad is most useful

More information

1 Generating functions for balls in boxes

1 Generating functions for balls in boxes Math 566 Fall 05 Some otes o geeratig fuctios Give a sequece a 0, a, a,..., a,..., a geeratig fuctio some way of represetig the sequece as a fuctio. There are may ways to do this, with the most commo ways

More information

Section 11.8: Power Series

Section 11.8: Power Series Sectio 11.8: Power Series 1. Power Series I this sectio, we cosider geeralizig the cocept of a series. Recall that a series is a ifiite sum of umbers a. We ca talk about whether or ot it coverges ad i

More information

Econ 325 Notes on Point Estimator and Confidence Interval 1 By Hiro Kasahara

Econ 325 Notes on Point Estimator and Confidence Interval 1 By Hiro Kasahara Poit Estimator Eco 325 Notes o Poit Estimator ad Cofidece Iterval 1 By Hiro Kasahara Parameter, Estimator, ad Estimate The ormal probability desity fuctio is fully characterized by two costats: populatio

More information

M1 for method for S xy. M1 for method for at least one of S xx or S yy. A1 for at least one of S xy, S xx, S yy correct. M1 for structure of r

M1 for method for S xy. M1 for method for at least one of S xx or S yy. A1 for at least one of S xy, S xx, S yy correct. M1 for structure of r Questio 1 (i) EITHER: 1 S xy = xy x y = 198.56 1 19.8 140.4 =.44 x x = 1411.66 1 19.8 = 15.657 1 S xx = y y = 1417.88 1 140.4 = 9.869 14 Sxy -.44 r = = SxxSyy 15.6579.869 = 0.76 1 S yy = 14 14 M1 for method

More information

MATH 320: Probability and Statistics 9. Estimation and Testing of Parameters. Readings: Pruim, Chapter 4

MATH 320: Probability and Statistics 9. Estimation and Testing of Parameters. Readings: Pruim, Chapter 4 MATH 30: Probability ad Statistics 9. Estimatio ad Testig of Parameters Estimatio ad Testig of Parameters We have bee dealig situatios i which we have full kowledge of the distributio of a radom variable.

More information

Access to the published version may require journal subscription. Published with permission from: Elsevier.

Access to the published version may require journal subscription. Published with permission from: Elsevier. This is a author produced versio of a paper published i Statistics ad Probability Letters. This paper has bee peer-reviewed, it does ot iclude the joural pagiatio. Citatio for the published paper: Forkma,

More information

Preponderantly increasing/decreasing data in regression analysis

Preponderantly increasing/decreasing data in regression analysis Croatia Operatioal Research Review 269 CRORR 7(2016), 269 276 Prepoderatly icreasig/decreasig data i regressio aalysis Darija Marković 1, 1 Departmet of Mathematics, J. J. Strossmayer Uiversity of Osijek,

More information

ANALYSIS OF EXPERIMENTAL ERRORS

ANALYSIS OF EXPERIMENTAL ERRORS ANALYSIS OF EXPERIMENTAL ERRORS All physical measuremets ecoutered i the verificatio of physics theories ad cocepts are subject to ucertaities that deped o the measurig istrumets used ad the coditios uder

More information

University of California, Los Angeles Department of Statistics. Practice problems - simple regression 2 - solutions

University of California, Los Angeles Department of Statistics. Practice problems - simple regression 2 - solutions Uiversity of Califoria, Los Ageles Departmet of Statistics Statistics 00C Istructor: Nicolas Christou EXERCISE Aswer the followig questios: Practice problems - simple regressio - solutios a Suppose y,

More information

ENGI 4421 Confidence Intervals (Two Samples) Page 12-01

ENGI 4421 Confidence Intervals (Two Samples) Page 12-01 ENGI 44 Cofidece Itervals (Two Samples) Page -0 Two Sample Cofidece Iterval for a Differece i Populatio Meas [Navidi sectios 5.4-5.7; Devore chapter 9] From the cetral limit theorem, we kow that, for sufficietly

More information

GUIDE FOR THE USE OF THE DECISION SUPPORT SYSTEM (DSS)*

GUIDE FOR THE USE OF THE DECISION SUPPORT SYSTEM (DSS)* GUIDE FOR THE USE OF THE DECISION SUPPORT SYSTEM (DSS)* *Note: I Frech SAD (Système d Aide à la Décisio) 1. Itroductio to the DSS Eightee statistical distributios are available i HYFRAN-PLUS software to

More information

STATISTICAL PROPERTIES OF LEAST SQUARES ESTIMATORS. Comments:

STATISTICAL PROPERTIES OF LEAST SQUARES ESTIMATORS. Comments: Recall: STATISTICAL PROPERTIES OF LEAST SQUARES ESTIMATORS Commets:. So far we have estimates of the parameters! 0 ad!, but have o idea how good these estimates are. Assumptio: E(Y x)! 0 +! x (liear coditioal

More information

Estimation of a population proportion March 23,

Estimation of a population proportion March 23, 1 Social Studies 201 Notes for March 23, 2005 Estimatio of a populatio proportio Sectio 8.5, p. 521. For the most part, we have dealt with meas ad stadard deviatios this semester. This sectio of the otes

More information

The standard deviation of the mean

The standard deviation of the mean Physics 6C Fall 20 The stadard deviatio of the mea These otes provide some clarificatio o the distictio betwee the stadard deviatio ad the stadard deviatio of the mea.. The sample mea ad variace Cosider

More information

THE SYSTEMATIC AND THE RANDOM. ERRORS - DUE TO ELEMENT TOLERANCES OF ELECTRICAL NETWORKS

THE SYSTEMATIC AND THE RANDOM. ERRORS - DUE TO ELEMENT TOLERANCES OF ELECTRICAL NETWORKS R775 Philips Res. Repts 26,414-423, 1971' THE SYSTEMATIC AND THE RANDOM. ERRORS - DUE TO ELEMENT TOLERANCES OF ELECTRICAL NETWORKS by H. W. HANNEMAN Abstract Usig the law of propagatio of errors, approximated

More information

MAT 271 Project: Partial Fractions for certain rational functions

MAT 271 Project: Partial Fractions for certain rational functions MAT 7 Project: Partial Fractios for certai ratioal fuctios Prerequisite kowledge: partial fractios from MAT 7, a very good commad of factorig ad complex umbers from Precalculus. To complete this project,

More information

Lecture 2: Monte Carlo Simulation

Lecture 2: Monte Carlo Simulation STAT/Q SCI 43: Itroductio to Resamplig ethods Sprig 27 Istructor: Ye-Chi Che Lecture 2: ote Carlo Simulatio 2 ote Carlo Itegratio Assume we wat to evaluate the followig itegratio: e x3 dx What ca we do?

More information

Lecture 7: Properties of Random Samples

Lecture 7: Properties of Random Samples Lecture 7: Properties of Radom Samples 1 Cotiued From Last Class Theorem 1.1. Let X 1, X,...X be a radom sample from a populatio with mea µ ad variace σ

More information

Problems from 9th edition of Probability and Statistical Inference by Hogg, Tanis and Zimmerman:

Problems from 9th edition of Probability and Statistical Inference by Hogg, Tanis and Zimmerman: Math 224 Fall 2017 Homework 4 Drew Armstrog Problems from 9th editio of Probability ad Statistical Iferece by Hogg, Tais ad Zimmerma: Sectio 2.3, Exercises 16(a,d),18. Sectio 2.4, Exercises 13, 14. Sectio

More information

Example: Find the SD of the set {x j } = {2, 4, 5, 8, 5, 11, 7}.

Example: Find the SD of the set {x j } = {2, 4, 5, 8, 5, 11, 7}. 1 (*) If a lot of the data is far from the mea, the may of the (x j x) 2 terms will be quite large, so the mea of these terms will be large ad the SD of the data will be large. (*) I particular, outliers

More information

Power and Type II Error

Power and Type II Error Statistical Methods I (EXST 7005) Page 57 Power ad Type II Error Sice we do't actually kow the value of the true mea (or we would't be hypothesizig somethig else), we caot kow i practice the type II error

More information

Discrete Mathematics for CS Spring 2008 David Wagner Note 22

Discrete Mathematics for CS Spring 2008 David Wagner Note 22 CS 70 Discrete Mathematics for CS Sprig 2008 David Wager Note 22 I.I.D. Radom Variables Estimatig the bias of a coi Questio: We wat to estimate the proportio p of Democrats i the US populatio, by takig

More information

Linear Regression Demystified

Linear Regression Demystified Liear Regressio Demystified Liear regressio is a importat subject i statistics. I elemetary statistics courses, formulae related to liear regressio are ofte stated without derivatio. This ote iteds to

More information

CHAPTER 10 INFINITE SEQUENCES AND SERIES

CHAPTER 10 INFINITE SEQUENCES AND SERIES CHAPTER 10 INFINITE SEQUENCES AND SERIES 10.1 Sequeces 10.2 Ifiite Series 10.3 The Itegral Tests 10.4 Compariso Tests 10.5 The Ratio ad Root Tests 10.6 Alteratig Series: Absolute ad Coditioal Covergece

More information

April 18, 2017 CONFIDENCE INTERVALS AND HYPOTHESIS TESTING, UNDERGRADUATE MATH 526 STYLE

April 18, 2017 CONFIDENCE INTERVALS AND HYPOTHESIS TESTING, UNDERGRADUATE MATH 526 STYLE April 18, 2017 CONFIDENCE INTERVALS AND HYPOTHESIS TESTING, UNDERGRADUATE MATH 526 STYLE TERRY SOO Abstract These otes are adapted from whe I taught Math 526 ad meat to give a quick itroductio to cofidece

More information

Output Analysis (2, Chapters 10 &11 Law)

Output Analysis (2, Chapters 10 &11 Law) B. Maddah ENMG 6 Simulatio Output Aalysis (, Chapters 10 &11 Law) Comparig alterative system cofiguratio Sice the output of a simulatio is radom, the comparig differet systems via simulatio should be doe

More information

The Random Walk For Dummies

The Random Walk For Dummies The Radom Walk For Dummies Richard A Mote Abstract We look at the priciples goverig the oe-dimesioal discrete radom walk First we review five basic cocepts of probability theory The we cosider the Beroulli

More information

Chapter 11 Output Analysis for a Single Model. Banks, Carson, Nelson & Nicol Discrete-Event System Simulation

Chapter 11 Output Analysis for a Single Model. Banks, Carson, Nelson & Nicol Discrete-Event System Simulation Chapter Output Aalysis for a Sigle Model Baks, Carso, Nelso & Nicol Discrete-Evet System Simulatio Error Estimatio If {,, } are ot statistically idepedet, the S / is a biased estimator of the true variace.

More information

A General Family of Estimators for Estimating Population Variance Using Known Value of Some Population Parameter(s)

A General Family of Estimators for Estimating Population Variance Using Known Value of Some Population Parameter(s) Rajesh Sigh, Pakaj Chauha, Nirmala Sawa School of Statistics, DAVV, Idore (M.P.), Idia Floreti Smaradache Uiversity of New Meico, USA A Geeral Family of Estimators for Estimatig Populatio Variace Usig

More information

17. Joint distributions of extreme order statistics Lehmann 5.1; Ferguson 15

17. Joint distributions of extreme order statistics Lehmann 5.1; Ferguson 15 17. Joit distributios of extreme order statistics Lehma 5.1; Ferguso 15 I Example 10., we derived the asymptotic distributio of the maximum from a radom sample from a uiform distributio. We did this usig

More information

Stat 421-SP2012 Interval Estimation Section

Stat 421-SP2012 Interval Estimation Section Stat 41-SP01 Iterval Estimatio Sectio 11.1-11. We ow uderstad (Chapter 10) how to fid poit estimators of a ukow parameter. o However, a poit estimate does ot provide ay iformatio about the ucertaity (possible

More information

Time series models 2007

Time series models 2007 Norwegia Uiversity of Sciece ad Techology Departmet of Mathematical Scieces Solutios to problem sheet 1, 2007 Exercise 1.1 a Let Sc = E[Y c 2 ]. The This gives Sc = EY 2 2cEY + c 2 ds dc = 2EY + 2c = 0

More information

The Riemann Zeta Function

The Riemann Zeta Function Physics 6A Witer 6 The Riema Zeta Fuctio I this ote, I will sketch some of the mai properties of the Riema zeta fuctio, ζ(x). For x >, we defie ζ(x) =, x >. () x = For x, this sum diverges. However, we

More information

Approximations to the Distribution of the Sample Correlation Coefficient

Approximations to the Distribution of the Sample Correlation Coefficient World Academy of Sciece Egieerig ad Techology Iteratioal Joural of Mathematical ad Computatioal Scieces Vol:5 No:4 0 Approximatios to the Distributio of the Sample Correlatio Coefficiet Joh N Haddad ad

More information

A Note on Box-Cox Quantile Regression Estimation of the Parameters of the Generalized Pareto Distribution

A Note on Box-Cox Quantile Regression Estimation of the Parameters of the Generalized Pareto Distribution A Note o Box-Cox Quatile Regressio Estimatio of the Parameters of the Geeralized Pareto Distributio JM va Zyl Abstract: Makig use of the quatile equatio, Box-Cox regressio ad Laplace distributed disturbaces,

More information

Discrete Mathematics for CS Spring 2007 Luca Trevisan Lecture 22

Discrete Mathematics for CS Spring 2007 Luca Trevisan Lecture 22 CS 70 Discrete Mathematics for CS Sprig 2007 Luca Trevisa Lecture 22 Aother Importat Distributio The Geometric Distributio Questio: A biased coi with Heads probability p is tossed repeatedly util the first

More information

Z ß cos x + si x R du We start with the substitutio u = si(x), so du = cos(x). The itegral becomes but +u we should chage the limits to go with the ew

Z ß cos x + si x R du We start with the substitutio u = si(x), so du = cos(x). The itegral becomes but +u we should chage the limits to go with the ew Problem ( poits) Evaluate the itegrals Z p x 9 x We ca draw a right triagle labeled this way x p x 9 From this we ca read off x = sec, so = sec ta, ad p x 9 = R ta. Puttig those pieces ito the itegralrwe

More information

Stat 139 Homework 7 Solutions, Fall 2015

Stat 139 Homework 7 Solutions, Fall 2015 Stat 139 Homework 7 Solutios, Fall 2015 Problem 1. I class we leared that the classical simple liear regressio model assumes the followig distributio of resposes: Y i = β 0 + β 1 X i + ɛ i, i = 1,...,,

More information

Activity 3: Length Measurements with the Four-Sided Meter Stick

Activity 3: Length Measurements with the Four-Sided Meter Stick Activity 3: Legth Measuremets with the Four-Sided Meter Stick OBJECTIVE: The purpose of this experimet is to study errors ad the propagatio of errors whe experimetal data derived usig a four-sided meter

More information

A NEW METHOD FOR CONSTRUCTING APPROXIMATE CONFIDENCE INTERVALS FOR M-ESTU1ATES. Dennis D. Boos

A NEW METHOD FOR CONSTRUCTING APPROXIMATE CONFIDENCE INTERVALS FOR M-ESTU1ATES. Dennis D. Boos .- A NEW METHOD FOR CONSTRUCTING APPROXIMATE CONFIDENCE INTERVALS FOR M-ESTU1ATES by Deis D. Boos Departmet of Statistics North Carolia State Uiversity Istitute of Statistics Mimeo Series #1198 September,

More information