AMS Computational Finance

Size: px
Start display at page:

Download "AMS Computational Finance"

Transcription

1 ams-514-lec-13-m.nb 1 AMS Computational Finance Workshop Robert J. Frey, Research Professor Stony Brook University, Applied Mathematics and Statistics frey@ams.sunysb.edu April 2006 Overview This workshop deals with a complete end-to-end portfolio analysis. Beginning with raw price histories of the current components of the Dow Jones Industrial Index and ending up the computation of the mean-variance frontier of efficient portfolios. Table of Contents I. Software A. Set Up B. Downloading Data From A URL C. History Data From Yahoo!Finance D. Factor Models E. Portfolio Optimization II. Assignment A. Getting The Data B. Estimating The Portfolio Statistics C. Building A 2-Factor Model D. Portfolio Optimizations I. Software This section contains the descriptions and source code for the software used in the assignment. Where applicable, the underlying mathematical theory is outlined as well.

2 ams-514-lec-13-m.nb 2 A. Set Up The following standard packages are required. In addition, the setting of spelling warnings is adjusted to avoid a large number of spurious warnings. << JLink` << Miscellaneous`Calendar` << Statistics`MultiDescriptiveStatistics` << Statistics`LinearRegression` << Graphics`Graphics` Off@General::spell1 B. Downloading Data From A URL This function is based on the technical note on the Wolfram Research website: Elli, Brendan, "Using Java to get Stock Data from the Internet into Mathematica": This function reads data from the specified URL, stores it in a temporary file and returns the name of the file for further processing.

3 ams-514-lec-13-m.nb 3 xgeturl@surl_stringd := JavaBlock@Module@ 8uBuffer, icount, sotmp, ourl, siurl<, InstallJava@ sotmp = OpenTemporary@DOSTextFormat -> False H* Create a temp file. *L ourl = JavaNew@"java.net.URL", surl H* $Failed if not valid. *L siurl = ourl ü openstream@ H* Create a Java input stream. *L If@siUrl === $Failed, Return@$FailedD ubuffer = JavaNew@"@B", 5000 H* 5000 is an arbitrary size. *L While@ H* Read data from the stream saving it to the temp file. *L HiCount = siurl ü read@ubufferdl > 0, WriteString@soTmp, FromCharacterCode@Take@Val@uBufferD, icountddd siurl ü close@ H* Close the stream. *L Close@soTmpD H* Close the temp file, which also returns the filename. *L D This is a useful, general purpose function that will load a text representation of a URL's content into Mathematica. It's most straightforward application is in instances in which the URL returns its data as conveniently formatted text, such as comma separated value. It can also be used to download a page in HTML format. The challenges of parsing such data from within Mathematica are greater; however, Mathematica's ability to parse XML text using its XML capabilities is extremely useful. C. History Data From Yahoo!Fianance This collection of routines are used to build a simple function which will download stock data from Yahoo!Finance Utility Functions The formatting of the data is perfectly readable but is hardly optimized for use within Mathematica. These functions work to reformat these data into something more useful. ü Regularizing field names We want to be able to regularize field names by squeezing out spaces and extraneous (non-letter and non-digit) characters. xregularizename = StringJoin ü Pick@#, HLetterQ@#D»» DigitQ@#DL & êü #D & ü Characters@#D &; ü Date formatting The dates are returned as strings, e.g., {2006, 3, 27} is "27-Mar-06". The following set of functions convert these strings into Mathematica date vectors. The first converts 3-letter month names to integers.

4 ams-514-lec-13-m.nb 4 xmonthnametointeger@ss_stringd := Which@ ss ã "Jan", 1, ss ã "Feb", 2, ss ã "Mar", 3, ss ã "Apr", 4, ss ã "May", 5, ss ã "Jun", 6, ss ã "Jul", 7, ss ã "Aug", 8, ss ã "Sep", 9, ss ã "Oct", 10, ss ã "Nov", 11, ss ã "Dec", 12, True, $Failed This function converts a 2-digit year into a four digit year. It makes a guest at what point to switch from 1900 to 2000 based on the current year. xcompleteyear@iyear_integerd := Module@ 8iDivide<, idivide = Round@100 FractionalPart@First@Date@DD ê 100.D Which@ iyear 1900, iyear, iyear 100, If@iYear idivide, iyear , iyear D, True, $Failed D This function ties the prior two functions together to convert a Yahoo formatted date to one more usable by Mathematica. Here's a demonstration. xyahoodatetomathematicadate@ss_stringd := Module@ 8vsS<, vss = StringSplit@sS, "-" 8xCompleteYear@ToExpression@vsSP3TDD, xmonthnametointeger@vssp2td, ToExpression@vsSP1TD< xyahoodatetomathematicadate@"27-mar-06"d 82006, 3, 27<

5 ams-514-lec-13-m.nb 5 Building a URL The example in the Technical Note referenced above uses an MSN site. For various reasons, we have decided to use Yahoo!Finance as our data source primarily because its split- and dividend-adjusted "Adj Close" is in keeping with CRSP standards. This function builds a URL which will download daily data in comma-separated-value (CSV) format. Unfortunately, we don't have documentation for this; we were able to infer the required arguments from using the interactive historic download facility on Yahoo!Finance. xyahoofinancehistoryurl@ ssymbol_string, 8iStartYear_Integer, istartmonth_integer, istartday_integer<, 8iEndYear_Integer, iendmonth_integer, iendday_integer< D := Module@ 8xF, ss<, xf@ss_d := StringTake@"0" <> ToString@sSD, -2 StringJoin@ " H* Main call *L "s=" <> ssymbol, H* Symbol string *L "&a=" <> xf@istartmonth - 1D, H* Start month - 1 *L "&b=" <> ToString@iStartDayD, H* Start day *L "&c=" <> ToString@iStartYearD, H* Start year *L "&d=" <> xf@iendmonth - 1D, H* End month - 1 *L "&e=" <> ToString@iEndDayD, H* End day *L "&f=" <> ToString@iEndYearD, H* End year *L "&g=d", H* Frequency: daily *L "&ignore=.csv" D Downloading, reading in and processing the data. Given the data specification, as above, this function passes it to xyahoofinancehistoryurl, uses the URL to download the data, reads the data into a local variable and performs some postprocessing to put it into a convenient form for subsequent processing in Mathematica. The function returns a 3-vector consisting of the original arguments, a list of field names, and a data matrix.

6 ams-514-lec-13-m.nb 6 xgetyahoofinancehistorydata@ssymbol_string, vistart_, viend_d := Module@ 8vvsBuffer, vvudata, vsfields<, vvsbuffer = Import@ xgeturl@ xyahoofinancehistoryurl@ssymbol, vistart, viendd D, "CSV" vsfields = xregularizename êü First@vvsBuffer vvudata = Rest@vvsBuffer vvudatapall, 1T = xyahoodatetomathematicadate êü vvudatapall, 1T; vvudata = Sort@vvuData, HFirst@#1D , 100, 1<L < HFirst@#2D , 100, 1<L & 88sSymbol, vistart, viend<, vsfields, vvudata< Here are the data for IBM for the first half of June, Note, only trading days are retrieved. xgetyahoofinancehistorydata@"ibm", 82005, 6, 1<, 82005, 6, 15<D 88IBM, 82005, 6, 1<, 82005, 6, 15<<, 8Date, Open, High, Low, Close, Volume, AdjClose<, , 6, 1<, 75.57, 77.5, 75.57, 76.84, , 76.28<, , 6, 2<, 76.75, 77.39, 76.68, 77.35, , 76.79<, , 6, 3<, 77.06, 77.1, 75.74, 75.79, , 75.24<, , 6, 6<, 75.8, 75.9, 74.92, 75., , 74.45<, , 6, 7<, 75., 76.09, 75., 75.04, , 74.49<, , 6, 8<, 75.04, 75.4, 74.63, 74.8, , 74.26<, , 6, 9<, 74.58, 75.47, 74.23, 74.93, , 74.38<, , 6, 10<, 74.75, 75.05, 74.1, 74.77, , 74.23<, , 6, 13<, 74.5, 75.93, 74.45, 75.05, , 74.5<, , 6, 14<, 75.05, 75.43, 74.73, 74.89, , 74.34<, , 6, 15<, 75.7, 76.5, 75.15, 76.3, , 75.74<<< D. Factor Models Theory Given an n µ n covariance matrix Q, the factor model of order k is based on the n µ k factor exposure matrix B and diagonal error matrix D such that Q > BB T + D Typically, k << n. The factor model above is valid for factors returns which are uncorrelated and of unit variance. This function computes a factor model of order iorder from the covariance matrix mncov. Termination occurs when either the cosine of the angle between successive estimates of the main diagonal of D is less than ntol or the iteration

7 ams-514-lec-13-m.nb 7 limit imaxiter is reached. The function returns the iteration count, "error" cosine, factor exposures and main diagonal of the error covariance. The vector d is the main diagonal of the error covariance D. The main iteration starts with B i.b i T = Q DiagonalMatrix[d i-1 ] The matrix B i is computed using the singular value decomposition of B i.b i T. For a factor model of order k, the first k components of the svd are used. A new estimate of the main diagonal of D can now be computed. d i = Tr[Q B i.b i T, List] This continues until an iteration limit is reached or the cosine between successive values of d falls at least at some tolerance t. Total@d i d i-1 D ÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅ "#################### Total@d 2 i D "################ Total@d ########## 2 i-1 D ÅÅÅÅÅÅÅÅÅÅÅÅ t The default tolerance is The process must also be initialized and a value of d 0 = ÅÅÅÅ 1 Tr[Q, List] seems to work 2 well in practice. Code xfitfactormodel@mncov_, iorder_, ntol_: H L, imaxiter_: 10D := ModuleA 8d, f, i, n, p, r<, d = Tr@mnCov, ListD ê 2.; i = 0; n = 0; WhileAHi < imaxiterl && Hn ntoll, p = d; r = mncov - DiagonalMatrix@d f = IFirst@#D. è!!!!!!!!!!! #P2T M &@ SingularValueDecomposition@r, iorderd d = Tr@mnCov - f.transpose@fd, List Total@p dd n = ÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅ è!!!!!!!!!!!!!!!!!!!!!! ÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅ Total@p 2 D è!!!!!!!!!!!!!!!!!!!!!! ; Total@d 2 D i++ E; 8i, n, f, d< E; E. Portfolio Optimization Theory Our standard form for mean-variance portfolio optimization will be allocate one unit of capital with no short sales allowed:

8 ams-514-lec-13-m.nb 8 naïve [l, Q, r] = max x 8 1 ÅÅÅÅ 2 xt Q x l r T x» x 0, T x = 1< If the covariance matrix above can be stated as a factor model then the variance portion of the objective function can be restated as follows. Q = B B T + D ï x T Q x = x T B B T x + x T D x We can next make a substitution in which a vector y represents the factor exposure of the portfolio. B T x = y ï x T B B T x + x T D x = y T y + x T D x This allows us to restate the original optimization in the following form. factor [l, B, D, r] = max i k j x y z y{ 9 ÅÅÅÅ 1 2 J x T y N J D 0 0 I N J x y N l rt x x 0, T x = 1, B T x = y= ƒ The advantage here is that the covariance matrix the Hessian of the objective function can be expressed as a diagonal matrix. This considerably simplifies many computations. The disadvantage is that we have added a large number of "accounting" constraints to get the factor exposures y right. Generally, the second form of the quadratic program requires much less memory and computation time when reasonable effort is taken to use the simplifications available, so despite the large number of constraints added, it usually a good idea to factor the covariance matrix. The runtime required in the first case increases cubicly in the size of the problem. In the second case, the runtime increases slightly worse than linear in the size of the problem. This means that large problems that can not be practically handled with a naïve covariance model are easily solved with an appropriate factorization. Other factorizations will do. For example, a Cholesky decomposition of Q = CC T could be used to define constraints C T x = y and the new formulation would be Cholesky [l, C, r] = max i k j x y z y{ 8 ÅÅÅÅ 1 2 yt y l r T x» x 0, T x = 1, C T x = y< Factor models have other benefits, however. The fact that they represent the covariance structure of the trading universe in a more parsimonious form than the full covariance matrix gives them greater numerical and statistical stability. Code Although it is not an ideal platform, the built-in Mathematica function NMinimize works quite well for small to medium sized portfolio optimization problems. Its practical use in this regard is considerably extended by employing the approach above. The first function below assumes that the covariance structure of the market can be represented by a factor model. The second assumes a naïve covariance matrix. Review NMinimize for additional insight into how they work. The inputs to the first function are: the risk tolerance l, a vector of strings giving the names of the instruments, a vector of strings giving the names of the factors, the mean vector r, the factor loading matrix B, and a vector representing the main diagonal of D. The inputs to the second function are: the risk tolerance l, a vector of strings giving the names of the instruments, the mean vector r and the covariance matrix Q. Both return their result in the same form: a 2-vector whose first element is a list containing the portfolio variance and mean and whose last element is a list containing the portfolio given in the same order as the description vector for the instruments. Although the same symbol, stdqpsolve, is used for both functions, Mathematica distinguishes between them based on the pattern of their arguments.

9 ams-514-lec-13-m.nb 9 stdqpsolve@ nrisktol_, vsinstrumentdesc_, vsfactordesc_, vninstrumentmean_, mnfactorloading_, vndiagonal_ D := Module@ 8cons, conscapital, consnonnegativity, port, obj, sol, vars, varsfactor, varsinstrument<, varsinstrument = Unique@# <> "$" & êü vsinstrumentdesc varsfactor = Unique@# <> "$" & êü vsfactordesc vars = Join@varsInstrument, varsfactor consfactorloading = H# ã 0 &L êü HHH#.varsInstrument &L êü HmnFactorLoading LL - varsfactorl; consnonnegativity = H# 0 &L êü varsinstrument; conscapital = 8Plus üü varsinstrument ã 1<; cons = And üü Join@consFactorLoading, consnonnegativity, conscapital obj = 0.5 HTotal@Join@HvarsInstrument 2 vndiagonall, varsfactor 2 DDL - nrisktol HvnInstrumentMean.varsInstrumentL; sol = NMinimize@8obj, cons<, vars, PrecisionGoal Ø 6 port = varsinstrument ê. solp2t; 88H#.#L &@port.mnfactord + Hport 2.vnDiagonalL, port.vnmean<, port< stdqpsolve@ nrisktol_, vsdesc_, vnmean_, mncov_ D := Module@ 8cons, port, obj, sol, vars<, vars = Unique@# <> "$" & êü vsdesc cons = And üü Join@8Plus üü vars ã 1<, H# 0 &L êü vars obj = 0.5 Hvars.mnCov.varsL - nrisktol HvnMean.varsL; sol = NMinimize@8obj, cons<, vars, PrecisionGoal Ø 6 port = vars ê. solp2t; 88port.mnCov.port, port.vnmean<, port< II. Assignment The assignment was to plot the efficient frontier for the current members of the Dow Jones Industrial Index based on the price history that was available in common. This was to be done using a 2-factor model first. These results were to be contrasted with using the full covariance matrix.

10 ams-514-lec-13-m.nb 10 A. Getting the Data Stock Universe The list of tickers of the Dow Jones Industrial Average. vsdowjonessymbol = 8"MMM", "AA", "MO", "AXP", "AIG", "T", "BA", "CAT", "C", "KO", "DD", "XOM", "GE", "GM", "HPQ", "HD", "HON", "INTC", "IBM", "JNJ", "JPM", "MCD", "MRK", "MSFT", "PFE", "PG", "UTX", "VZ", "WMT", "DIS"< 8MMM, AA, MO, AXP, AIG, T, BA, CAT, C, KO, DD, XOM, GE, GM, HPQ, HD, HON, INTC, IBM, JNJ, JPM, MCD, MRK, MSFT, PFE, PG, UTX, VZ, WMT, DIS< Downloading Raw Daily History Data Ultimately, we want only month ending dates and, as was established in earlier classes, it appears that we can only as far back as {1986, 7, 9}. There also isn't much point including anything past {2006, 3, 31} since we're still in April. audataabstract = xgetyahoofinancehistorydata@#, 81986, 7, 9<, 82006, 3, 31<D & êü vsdowjonessymbol; We only need the Date and AdjClose fields, so we'll restructure dataabstract to that effect. audataabstract = 8#P1T, #P2, 81, 7<T, #P3, All, 81, 7<T< & êü audataabstract; We only need the month ending dates. The Split function can be used to split a list into sublists of "same" elements. Here we employ Split with a function that defines two entries as the same if the month of their dates are the same. Applied to a single element of the top-level dataabstract list this returns sublists of month data. We then simply pull out the last element of each sublist as the last trading day of the month. Pulling Out Month Ending Dates aumonthlyabstract = 8#P1T, #P2T, Last êü Split@#P3T, H#1P1, 2T ã #2P1, 2TL &D< & êü audataabstract; We want to be very careful and check that all of the dates across all of the stocks are equal. They are. SameQ üü H#P3, All, 1T & êü aumonthlyabstractl True

11 ams-514-lec-13-m.nb 11 Finally, we build a returns data structure consisting of three elements, The first is a list of dates as row names. The second is a list of tickers as column nams. The third is a matrix of returns over the dates and tickers. Building A Return Data Structure vureturns = 9 Rest@auMonthlyAbstractP1, 3, All, 1TD, aumonthlyabstractpall, 1, 1T, Rest@#D ÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅ Most@#D - 1 &@ Transpose@#P3, All, 2T & êü aumonthlyabstractd D =; Dimensions êü vureturns 88236, 3<, 830<, 8236, 30<< B. Estimating the Portfolio Statistics The vector of expected returns is vnmean = Mean@vuReturnsP3T Dimensions@vnMeanD 830< The covariance matrix of returns is mncovariance = CovarianceMatrix@vuReturnsP3T Dimensions@mnCovarianceD 830, 30< C. Building A 2-Factor Model Estimating The Model First, we estimate the factor model of order 2.

12 ams-514-lec-13-m.nb 12 8iIter, nfit, mnfactor, vndiag< = xfitfactormodel@mncovariance, 2 8iIter, nfit< 86, 1.< Then, we construct the approximate covariance matrix. mnapproxcov = HmnFactor.Transpose@mnFactorDL + DiagonalMatrix@vnDiag Evaluating The Model We generate a sample of portfolio allocation vectors x = {x 1, x 1,, x i,, x n } such that x i 0 and T x ã 1. For each element of the sample we compute the portfolio standard deviation that we would expect given the factor model and original covariance, respectively. mnsample = 9 è!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! #.mnapproxcov.#, è!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! #.mncovariance.#!!!!!!!!! = & êü TableA # ÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅ &@Table@Random@D, 830<DD, Total@#D 81000< E; Here is a plot of the factor model versus the full covarinace. The red line represents the line y = x.

13 ams-514-lec-13-m.nb 13 mnsample, PlotRange Ø All, PlotStyle Ø Frame Ø True, FrameLabel Ø 8 "Factor Model", "Full Covariance", StyleForm@"Portfolio s Comparison", FontSize Ø 16D, "" <, ImageSize Ø 400 D, Plot@ x, 8x, Min@Flatten@mnSampleDD, Max@Flatten@mnSampleDD<, PlotStyle Ø 8RGBColor@1, 0, 0D, Thickness@0.005D< D < Portfolio s Comparison F ull Covarianc e Factor Model The next plot is a histogram of the ratio between the factor model and full covariance standard deviation.

14 ams-514-lec-13-m.nb 14 HistogramA è!!!!!!!!!!!!!!!!!!!!!!!! #P1T ê #P2T & êü mnsample, PlotLabel Ø "ApproxêFull Portfolio Variance", FontSize Ø 16 D, ImageSize Ø 400 E; 150 ApproxêFull Portfolio Variance A 30 µ 30 covariance matrix has n(n + 1)/2 = 465 unique parameters (allowing for symmetry). The 2-factor model contains (k n) + n = (2 µ 30) + 30 = 90 parameters. When dealing with larger stock universes the differences are even greater. For example, if we had been working with the S&P 500, then a 2-factor model requires 1,500 parameters while the full covariance matrix contains 125,250 parameters. Thus, the factor model would contain 1.2% of the parameters of a naïve variance model. D. Portfolio Optimizations Runtime Comparison The relative runtime improvement of the optimization based on the factor model versus that of one based on the full covariance matrix varies depending upon the geometry of the problem at hand; however, even for a small problem such as this it is quite significant.

15 ams-514-lec-13-m.nb 15 vsdowjonessymbol, 8"Factor1", "Factor2"<, vnmean, mnfactor, vndiagd D Second, , <, 80., 0., , 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., , 0., , 0., 0., 0., 0., 0., , 0., 0., 0., 0., 0., 0.<<< Timing@ stdqpsolve@0.5, vsdowjonessymbol, vnmean, mncovarianced D Second, , <, 80., µ 10-8, , 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., , 0., , 0., 0., 0., 0., 0., , 0., 0., 0., 0., 0., 0.<<< There are small differences in the allocations, but the portfolio variance and mean are quite close to one another. Efficient Frontier Using A Factor Model After a little offline experimentation, the following list of risk tolerances appear to reasonably span the risk space from l = 0 (minimum variance, i.e., only risk considered) to l > 4 (all capital allocated to the single asset with the highest return, i.e., only expected return considered). vnrisktol = Prepend@2. Range@-4,2,0.5D, 0.D 80., , , 0.125, , 0.25, , 0.5, , 1., , 2., , 4.< We can now run through a series of optimizations at various risk levels. vusolutions = stdqpsolve@#, vsdowjonessymbol, 8"Factor1", "Factor2"<, vnmean, mnfactor, vndiagd & êü vnrisktol; Pulling out the efficient frontier is straightforward. mnefficientfrontier = First êü vusolutions; The efficient frontier is plotted below.

16 ams-514-lec-13-m.nb 16 mnefficientfrontier, Frame Ø True, PlotJoined Ø True, PlotRange Ø 880, 0.014<, 80, 0.035<<, ImageSize Ø Efficient Frontier Using Naïve Covariance Given the difference in runtimes, a coarser set of risk tolerances seems reasonable. vnalttol = Prepend@2. Range@-4,2,2D, 0.D 80., , 0.25, 1, 4.< As before we run through a series of optimizations at various risk levels. vualtsolutions = stdqpsolve@#, vsdowjonessymbol, vnmean, mncovarianced & êü vnalttol; And, again, pull out the efficient frontier from the results. mnaltefficientfrontier = First êü vualtsolutions;

17 ams-514-lec-13-m.nb 17 Here we plot the results using the naïve covariance approach as points against the efficient frontier computed using a factor model. There are small differences but the results are essentially the same. DisplayTogether@8 ListPlot@ mnefficientfrontier, Frame Ø True, PlotJoined Ø True, PlotRange Ø 880, 0.014<, 80, 0.035<<, ImageSize Ø 400 D, ListPlot@ mnaltefficientfrontier, PlotRange Ø 880, 0.014<, 80, 0.035<<, PlotStyle Ø 8PointSize@0.015D, RGBColor@1, 0, 0D< D <

DYNAMIC VS STATIC AUTOREGRESSIVE MODELS FOR FORECASTING TIME SERIES

DYNAMIC VS STATIC AUTOREGRESSIVE MODELS FOR FORECASTING TIME SERIES DYNAMIC VS STATIC AUTOREGRESSIVE MODELS FOR FORECASTING TIME SERIES Chris Xie Polytechnic Institute New York University (NYU), NY chris.xie@toprenergy.com Phone: 905-93-0577 June, 008 Electronic copy available

More information

AMS Foundations

AMS Foundations ams-5-sol-04-m.nb AMS 5 - Foundations Factor Models Robert J. Frey Research Professor Stony Brook University, Applied Mathematics and Statistics frey@ams.sunysb.edu Exercises for Class 3. The Chapters

More information

Multivariate elliptically contoured stable distributions: theory and estimation

Multivariate elliptically contoured stable distributions: theory and estimation Multivariate elliptically contoured stable distributions: theory and estimation John P. Nolan American University Revised 31 October 6 Abstract Mulitvariate stable distributions with elliptical contours

More information

GAMINGRE 8/1/ of 7

GAMINGRE 8/1/ of 7 FYE 09/30/92 JULY 92 0.00 254,550.00 0.00 0 0 0 0 0 0 0 0 0 254,550.00 0.00 0.00 0.00 0.00 254,550.00 AUG 10,616,710.31 5,299.95 845,656.83 84,565.68 61,084.86 23,480.82 339,734.73 135,893.89 67,946.95

More information

ESTIMATION AND FORECASTING OF LARGE REALIZED COVARIANCE MATRICES AND PORTFOLIO CHOICE LAURENT A. F. CALLOT

ESTIMATION AND FORECASTING OF LARGE REALIZED COVARIANCE MATRICES AND PORTFOLIO CHOICE LAURENT A. F. CALLOT ESTIMATION AND FORECASTING OF LARGE REALIZED COVARIANCE MATRICES AND PORTFOLIO CHOICE LAURENT A. F. CALLOT VU University Amsterdam, The Netherlands, CREATES, and the Tinbergen Institute. ANDERS B. KOCK

More information

Estimation and Forecasting of Large Realized Covariance Matrices and Portfolio Choice. Laurent A. F. Callot, Anders B. Kock and Marcelo C.

Estimation and Forecasting of Large Realized Covariance Matrices and Portfolio Choice. Laurent A. F. Callot, Anders B. Kock and Marcelo C. Estimation and Forecasting of Large Realized Covariance Matrices and Portfolio Choice Laurent A. F. Callot, Anders B. Kock and Marcelo C. Medeiros CREATES Research Paper 2014-42 Department of Economics

More information

Chapter 1 Handout: Descriptive Statistics

Chapter 1 Handout: Descriptive Statistics Preview Chapter 1 Handout: Descriptive Statistics Describing a Single Data Variable o Introduction to Distributions o Measure of the Distribution Center: Mean (Average) o Measures of the Distribution Spread:

More information

Bayesian Stochastic Volatility (SV) Model with non-gaussian Errors

Bayesian Stochastic Volatility (SV) Model with non-gaussian Errors rrors Bayesian Stochastic Volatility (SV) Model with non-gaussian Errors Seokwoo Lee 1, Hedibert F. Lopes 2 1 Department of Statistics 2 Graduate School of Business University of Chicago May 19, 2008 rrors

More information

Stat 206, Week 6: Factor Analysis

Stat 206, Week 6: Factor Analysis Stat 206, Week 6: Factor Analysis James Johndrow 2016-09-24 Introduction Factor analysis aims to explain correlation between a large set (p) of variables in terms of a smaller number (m) of underlying

More information

Technical note on seasonal adjustment for M0

Technical note on seasonal adjustment for M0 Technical note on seasonal adjustment for M0 July 1, 2013 Contents 1 M0 2 2 Steps in the seasonal adjustment procedure 3 2.1 Pre-adjustment analysis............................... 3 2.2 Seasonal adjustment.................................

More information

Technical note on seasonal adjustment for Capital goods imports

Technical note on seasonal adjustment for Capital goods imports Technical note on seasonal adjustment for Capital goods imports July 1, 2013 Contents 1 Capital goods imports 2 1.1 Additive versus multiplicative seasonality..................... 2 2 Steps in the seasonal

More information

Forecasting the Canadian Dollar Exchange Rate Wissam Saleh & Pablo Navarro

Forecasting the Canadian Dollar Exchange Rate Wissam Saleh & Pablo Navarro Forecasting the Canadian Dollar Exchange Rate Wissam Saleh & Pablo Navarro Research Question: What variables effect the Canadian/US exchange rate? Do energy prices have an effect on the Canadian/US exchange

More information

Jayalath Ekanayake Jonas Tappolet Harald Gall Abraham Bernstein. Time variance and defect prediction in software projects: additional figures

Jayalath Ekanayake Jonas Tappolet Harald Gall Abraham Bernstein. Time variance and defect prediction in software projects: additional figures Jayalath Ekanayake Jonas Tappolet Harald Gall Abraham Bernstein TECHNICAL REPORT No. IFI-2.4 Time variance and defect prediction in software projects: additional figures 2 University of Zurich Department

More information

Reconstructing an economic space from a market metric

Reconstructing an economic space from a market metric arxiv:cond-mat/28v [cond-mat.stat-mech] 6 Nov 22 Reconstructing an economic space from a market metric R. Vilela Mendes Grupo de Física Matemática, Av. Gama Pinto 2, 699 Lisboa Codex, Portugal (vilela@cii.fc.ul.pt)

More information

Multivariate Distributions

Multivariate Distributions IEOR E4602: Quantitative Risk Management Spring 2016 c 2016 by Martin Haugh Multivariate Distributions We will study multivariate distributions in these notes, focusing 1 in particular on multivariate

More information

CENTRE FOR ECONOMETRIC ANALYSIS

CENTRE FOR ECONOMETRIC ANALYSIS CENTRE FOR ECONOMETRIC ANALYSIS CEA@Cass http://www.cass.city.ac.uk/cea/index.html Cass Business School Faculty of Finance 106 Bunhill Row London EC1Y 8TZ Co-features in Finance: Co-arrivals and Co-jumps

More information

ENGINE SERIAL NUMBERS

ENGINE SERIAL NUMBERS ENGINE SERIAL NUMBERS The engine number was also the serial number of the car. Engines were numbered when they were completed, and for the most part went into a chassis within a day or so. However, some

More information

Introduction to S+FinMetrics

Introduction to S+FinMetrics 2008 An Introduction to S+FinMetrics This workshop will unleash the power of S+FinMetrics module for the econometric modeling and prediction of economic and financial time series. S+FinMetrics library

More information

A Gradual Non-Convexation Method for Minimizing VaR

A Gradual Non-Convexation Method for Minimizing VaR A Gradual Non-Convexation Method for Minimizing VaR Jiong Xi Thomas F. Coleman Yuying Li Aditya Tayal July 23, 2012 Abstract Given a finite set of m scenarios, computing a portfolio with the minimium Value-at-Risk

More information

Forecasting using R. Rob J Hyndman. 1.3 Seasonality and trends. Forecasting using R 1

Forecasting using R. Rob J Hyndman. 1.3 Seasonality and trends. Forecasting using R 1 Forecasting using R Rob J Hyndman 1.3 Seasonality and trends Forecasting using R 1 Outline 1 Time series components 2 STL decomposition 3 Forecasting and decomposition 4 Lab session 5 Forecasting using

More information

STOCK CYCLES FORECAST

STOCK CYCLES FORECAST Written and Published by: Michael S. Jenkins STOCK CYCLES FORECAST P.O. Box 652 Cathedral Station PO, New York, N.Y. 10025-9998 WWW.StockCyclesForecast.com Volume 28 Issue 8 October 19, 2012 Dow 13,413

More information

Mr. XYZ. Stock Market Trading and Investment Astrology Report. Report Duration: 12 months. Type: Both Stocks and Option. Date: Apr 12, 2011

Mr. XYZ. Stock Market Trading and Investment Astrology Report. Report Duration: 12 months. Type: Both Stocks and Option. Date: Apr 12, 2011 Mr. XYZ Stock Market Trading and Investment Astrology Report Report Duration: 12 months Type: Both Stocks and Option Date: Apr 12, 2011 KT Astrologer Website: http://www.softwareandfinance.com/magazine/astrology/kt_astrologer.php

More information

Time series and Forecasting

Time series and Forecasting Chapter 2 Time series and Forecasting 2.1 Introduction Data are frequently recorded at regular time intervals, for instance, daily stock market indices, the monthly rate of inflation or annual profit figures.

More information

STATISTICAL FORECASTING and SEASONALITY (M. E. Ippolito; )

STATISTICAL FORECASTING and SEASONALITY (M. E. Ippolito; ) STATISTICAL FORECASTING and SEASONALITY (M. E. Ippolito; 10-6-13) PART I OVERVIEW The following discussion expands upon exponential smoothing and seasonality as presented in Chapter 11, Forecasting, in

More information

MSE 310/ECE 340. Instructor: Bill Knowlton. Initialize data Clear somedata, someplot. Place data in list somedata

MSE 310/ECE 340. Instructor: Bill Knowlton. Initialize data Clear somedata, someplot. Place data in list somedata MSE 3/ECE 340 Instructor: Bill Knowlton Examples of Plotting Functions ü Example 1: Using the command ListPlot[ ] to plot data in a scatter plot In[1]:= Initialize data Clear somedata someplot Place data

More information

chapter 11 ALGEBRAIC SYSTEMS GOALS

chapter 11 ALGEBRAIC SYSTEMS GOALS chapter 11 ALGEBRAIC SYSTEMS GOALS The primary goal of this chapter is to make the reader aware of what an algebraic system is and how algebraic systems can be studied at different levels of abstraction.

More information

Chapter 3 ANALYSIS OF RESPONSE PROFILES

Chapter 3 ANALYSIS OF RESPONSE PROFILES Chapter 3 ANALYSIS OF RESPONSE PROFILES 78 31 Introduction In this chapter we present a method for analysing longitudinal data that imposes minimal structure or restrictions on the mean responses over

More information

In this activity, students will compare weather data from to determine if there is a warming trend in their community.

In this activity, students will compare weather data from to determine if there is a warming trend in their community. Overview: In this activity, students will compare weather data from 1910-2000 to determine if there is a warming trend in their community. Objectives: The student will: use the Internet to locate scientific

More information

arxiv: v1 [q-fin.pm] 24 Apr 2010

arxiv: v1 [q-fin.pm] 24 Apr 2010 When do improved covariance matrix estimators enhance portfolio optimization? An empirical comparative study of nine estimators Ester Pantaleo, 1 Michele Tumminello, 2 Fabrizio Lillo, 2, 3 and Rosario

More information

Sales Analysis User Manual

Sales Analysis User Manual Sales Analysis User Manual Confidential Information This document contains proprietary and valuable, confidential trade secret information of APPX Software, Inc., Richmond, Virginia Notice of Authorship

More information

ECEN 651: Microprogrammed Control of Digital Systems Department of Electrical and Computer Engineering Texas A&M University

ECEN 651: Microprogrammed Control of Digital Systems Department of Electrical and Computer Engineering Texas A&M University ECEN 651: Microprogrammed Control of Digital Systems Department of Electrical and Computer Engineering Texas A&M University Prof. Mi Lu TA: Ehsan Rohani Laboratory Exercise #4 MIPS Assembly and Simulation

More information

Ch3. TRENDS. Time Series Analysis

Ch3. TRENDS. Time Series Analysis 3.1 Deterministic Versus Stochastic Trends The simulated random walk in Exhibit 2.1 shows a upward trend. However, it is caused by a strong correlation between the series at nearby time points. The true

More information

DEPARTMENT OF THE ARMY MILITARY SURFACE DEPLOYMENT AND DISTRIBUTION COMMAND (SDDC) 1 SOLDIER WAY SCOTT AFB, IL 62225

DEPARTMENT OF THE ARMY MILITARY SURFACE DEPLOYMENT AND DISTRIBUTION COMMAND (SDDC) 1 SOLDIER WAY SCOTT AFB, IL 62225 DEPARTMENT OF THE ARMY MILITARY SURFACE DEPLOYMENT AND DISTRIBUTION COMMAND (SDDC) 1 SOLDIER WAY SCOTT AFB, IL 62225 SDDC Operations Special Requirements Branch 1 Soldier Way Scott AFB, IL 62225 April

More information

Quantum Mechanics using Matrix Methods

Quantum Mechanics using Matrix Methods Quantum Mechanics using Matrix Methods Introduction and the simple harmonic oscillator In this notebook we study some problems in quantum mechanics using matrix methods. We know that we can solve quantum

More information

May 2017 Security Investment Report City of Lawrence, Kansas

May 2017 Security Investment Report City of Lawrence, Kansas May 2017 Security Investment Report TABLE OF CONTENTS CONSOLIDATED REPORTS BY PORTFOLIO 1. PORTFOLIO SUMMARY 1. PORTFOLIO SUMMARY 2. FIXED INCOME PORTFOLIO 2. FIXED INCOME PORTFOLIO 3. PERFORMANCE a. MONTHLY

More information

Markowitz Efficient Portfolio Frontier as Least-Norm Analytic Solution to Underdetermined Equations

Markowitz Efficient Portfolio Frontier as Least-Norm Analytic Solution to Underdetermined Equations Markowitz Efficient Portfolio Frontier as Least-Norm Analytic Solution to Underdetermined Equations Sahand Rabbani Introduction Modern portfolio theory deals in part with the efficient allocation of investments

More information

Lecture Prepared By: Mohammad Kamrul Arefin Lecturer, School of Business, North South University

Lecture Prepared By: Mohammad Kamrul Arefin Lecturer, School of Business, North South University Lecture 15 20 Prepared By: Mohammad Kamrul Arefin Lecturer, School of Business, North South University Modeling for Time Series Forecasting Forecasting is a necessary input to planning, whether in business,

More information

6.079/6.975 S. Boyd & P. Parrilo December 10 11, Final exam

6.079/6.975 S. Boyd & P. Parrilo December 10 11, Final exam 6.079/6.975 S. Boyd & P. Parrilo December 10 11, 2009. Final exam This is a 24 hour take-home final exam. Please turn it in to Professor Stephen Boyd, (Stata Center), on Friday December 11, at 5PM (or

More information

Enhancements to the CTBTO IDC radionuclide processing pipeline for particulate samples achieving significant improvement of automatic products

Enhancements to the CTBTO IDC radionuclide processing pipeline for particulate samples achieving significant improvement of automatic products Enhancements to the CTBTO IDC radionuclide processing pipeline for particulate samples achieving significant improvement of automatic products Hakim Gheddou, Martin Kalinowski, Elena Tomuta International

More information

Time Series Analysis

Time Series Analysis Time Series Analysis A time series is a sequence of observations made: 1) over a continuous time interval, 2) of successive measurements across that interval, 3) using equal spacing between consecutive

More information

Concepts of Approximate Error: Approximate Error, Absolute Approximate Error, Relative Approximate Error, and Absolute Relative Approximate Error

Concepts of Approximate Error: Approximate Error, Absolute Approximate Error, Relative Approximate Error, and Absolute Relative Approximate Error Concepts of Approximate Error: Approximate Error, Absolute Approximate Error, Relative Approximate Error, and Absolute Relative Approximate Error Sean Rodby, Luke Snyder, Autar Kaw University of South

More information

Weak Constraints 4D-Var

Weak Constraints 4D-Var Weak Constraints 4D-Var Yannick Trémolet ECMWF Training Course - Data Assimilation May 1, 2012 Yannick Trémolet Weak Constraints 4D-Var May 1, 2012 1 / 30 Outline 1 Introduction 2 The Maximum Likelihood

More information

Professor Wiston Adrián RISSO, PhD Institute of Economics (IECON), University of the Republic (Uruguay)

Professor Wiston Adrián RISSO, PhD   Institute of Economics (IECON), University of the Republic (Uruguay) Professor Wiston Adrián RISSO, PhD E-mail: arisso@iecon.ccee.edu.uy Institute of Economics (IECON), University of the Republic (Uruguay) A FIRST APPROACH ON TESTING NON-CAUSALITY WITH SYMBOLIC TIME SERIES

More information

Determine the trend for time series data

Determine the trend for time series data Extra Online Questions Determine the trend for time series data Covers AS 90641 (Statistics and Modelling 3.1) Scholarship Statistics and Modelling Chapter 1 Essent ial exam notes Time series 1. The value

More information

Annual Average NYMEX Strip Comparison 7/03/2017

Annual Average NYMEX Strip Comparison 7/03/2017 Annual Average NYMEX Strip Comparison 7/03/2017 To Year to Year Oil Price Deck ($/bbl) change Year change 7/3/2017 6/1/2017 5/1/2017 4/3/2017 3/1/2017 2/1/2017-2.7% 2017 Average -10.4% 47.52 48.84 49.58

More information

Package tawny. R topics documented: August 29, Type Package

Package tawny. R topics documented: August 29, Type Package Type Package Package tawny August 29, 2016 Title Clean Covariance Matrices Using Random Matrix Theory and Shrinkage Estimators for Portfolio Optimization Version 2.1.6 Depends R (>= 3.0.0) Imports lambda.r

More information

Web Appendix to Multivariate High-Frequency-Based Volatility (HEAVY) Models

Web Appendix to Multivariate High-Frequency-Based Volatility (HEAVY) Models Web Appendix to Multivariate High-Frequency-Based Volatility (HEAVY) Models Diaa Noureldin Department of Economics, University of Oxford, & Oxford-Man Institute, Eagle House, Walton Well Road, Oxford OX

More information

A Poor Example of a Project - But Gives Some Direction

A Poor Example of a Project - But Gives Some Direction A Poor Example of a Project - But Gives Some Direction Bill Knowlton Materials Science & Engineering Electrical & Computer Engineering Boise State Unversity The idea behind this example is two-fold: 1.

More information

Summary statistics. G.S. Questa, L. Trapani. MSc Induction - Summary statistics 1

Summary statistics. G.S. Questa, L. Trapani. MSc Induction - Summary statistics 1 Summary statistics 1. Visualize data 2. Mean, median, mode and percentiles, variance, standard deviation 3. Frequency distribution. Skewness 4. Covariance and correlation 5. Autocorrelation MSc Induction

More information

One-Way Repeated Measures Contrasts

One-Way Repeated Measures Contrasts Chapter 44 One-Way Repeated easures Contrasts Introduction This module calculates the power of a test of a contrast among the means in a one-way repeated measures design using either the multivariate test

More information

In Centre, Online Classroom Live and Online Classroom Programme Prices

In Centre, Online Classroom Live and Online Classroom Programme Prices In Centre, and Online Classroom Programme Prices In Centre Online Classroom Foundation Certificate Bookkeeping Transactions 430 325 300 Bookkeeping Controls 320 245 225 Elements of Costing 320 245 225

More information

= observed volume on day l for bin j = base volume in jth bin, and = residual error, assumed independent with mean zero.

= observed volume on day l for bin j = base volume in jth bin, and = residual error, assumed independent with mean zero. QB research September 4, 06 Page -Minute Bin Volume Forecast Model Overview In response to strong client demand, Quantitative Brokers (QB) has developed a new algorithm called Closer that specifically

More information

Modeling by the nonlinear stochastic differential equation of the power-law distribution of extreme events in the financial systems

Modeling by the nonlinear stochastic differential equation of the power-law distribution of extreme events in the financial systems Modeling by the nonlinear stochastic differential equation of the power-law distribution of extreme events in the financial systems Bronislovas Kaulakys with Miglius Alaburda, Vygintas Gontis, A. Kononovicius

More information

Statistical Models for Rainfall with Applications to Index Insura

Statistical Models for Rainfall with Applications to Index Insura Statistical Models for Rainfall with Applications to April 21, 2008 Overview The idea: Insure farmers against the risk of crop failure, like drought, instead of crop failure itself. It reduces moral hazard

More information

FEB DASHBOARD FEB JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC

FEB DASHBOARD FEB JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC Positive Response Compliance 215 Compliant 215 Non-Compliant 216 Compliant 216 Non-Compliant 1% 87% 96% 86% 96% 88% 89% 89% 88% 86% 92% 93% 94% 96% 94% 8% 6% 4% 2% 13% 4% 14% 4% 12% 11% 11% 12% JAN MAR

More information

ACCA Interactive Timetable

ACCA Interactive Timetable ACCA Interactive Timetable 2018 Professional information last updated 4 April 2018 v3.1 Please note: Information and dates in this timetable are subject to change. How the 4 exam sittings can work for

More information

Properties of Context-Free Languages

Properties of Context-Free Languages Properties of Context-Free Languages Seungjin Choi Department of Computer Science and Engineering Pohang University of Science and Technology 77 Cheongam-ro, Nam-gu, Pohang 37673, Korea seungjin@postech.ac.kr

More information

8.6 Recursion and Computer Algebra Systems

8.6 Recursion and Computer Algebra Systems 8.6 Recursion and Computer Algebra Systems There is frequently debate as to if how and when computers should be introduced teaching a topic such as recurrence relations. We have chosen to append this information

More information

The Instability of Correlations: Measurement and the Implications for Market Risk

The Instability of Correlations: Measurement and the Implications for Market Risk The Instability of Correlations: Measurement and the Implications for Market Risk Prof. Massimo Guidolin 20254 Advanced Quantitative Methods for Asset Pricing and Structuring Winter/Spring 2018 Threshold

More information

Tutorial on Principal Component Analysis

Tutorial on Principal Component Analysis Tutorial on Principal Component Analysis Copyright c 1997, 2003 Javier R. Movellan. This is an open source document. Permission is granted to copy, distribute and/or modify this document under the terms

More information

Trading Strategies using R

Trading Strategies using R Trading Strategies using R for the holy grail Eran Raviv Econometric Institute - Erasmus University, http://eranraviv.com April 02, 2012 Outline for section 1 1 introduction 2 3 4 1 5 9 13 17 21 25 29

More information

BUSI 460 Suggested Answers to Selected Review and Discussion Questions Lesson 7

BUSI 460 Suggested Answers to Selected Review and Discussion Questions Lesson 7 BUSI 460 Suggested Answers to Selected Review and Discussion Questions Lesson 7 1. The definitions follow: (a) Time series: Time series data, also known as a data series, consists of observations on a

More information

Climatography of the United States No

Climatography of the United States No Climate Division: AK 5 NWS Call Sign: ANC Month (1) Min (2) Month(1) Extremes Lowest (2) Temperature ( F) Lowest Month(1) Degree s (1) Base Temp 65 Heating Cooling 90 Number of s (3) Jan 22.2 9.3 15.8

More information

The xmacis Userʼs Guide. Keith L. Eggleston Northeast Regional Climate Center Cornell University Ithaca, NY

The xmacis Userʼs Guide. Keith L. Eggleston Northeast Regional Climate Center Cornell University Ithaca, NY The xmacis Userʼs Guide Keith L. Eggleston Northeast Regional Climate Center Cornell University Ithaca, NY September 22, 2004 Updated September 9, 2008 The xmacis Userʼs Guide The xmacis program consists

More information

2.1 Inductive Reasoning Ojectives: I CAN use patterns to make conjectures. I CAN disprove geometric conjectures using counterexamples.

2.1 Inductive Reasoning Ojectives: I CAN use patterns to make conjectures. I CAN disprove geometric conjectures using counterexamples. 2.1 Inductive Reasoning Ojectives: I CAN use patterns to make conjectures. I CAN disprove geometric conjectures using counterexamples. 1 Inductive Reasoning Most learning occurs through inductive reasoning,

More information

Applied Time Series Notes ( 1) Dates. ñ Internal: # days since Jan 1, ñ Need format for reading, one for writing

Applied Time Series Notes ( 1) Dates. ñ Internal: # days since Jan 1, ñ Need format for reading, one for writing Applied Time Series Notes ( 1) Dates ñ Internal: # days since Jan 1, 1960 ñ Need format for reading, one for writing ñ Often DATE is ID variable (extrapolates) ñ Program has lots of examples: options ls=76

More information

Objectives. Materials

Objectives. Materials . Objectives Activity 13 To graphically represent and analyze climate data To use linear regressions to understand the relationship between temperatures as measured in the Fahrenheit and Celsius scale

More information

SYSTEM BRIEF DAILY SUMMARY

SYSTEM BRIEF DAILY SUMMARY SYSTEM BRIEF DAILY SUMMARY * ANNUAL MaxTemp NEL (MWH) Hr Ending Hr Ending LOAD (PEAK HOURS 7:00 AM TO 10:00 PM MON-SAT) ENERGY (MWH) INCREMENTAL COST DAY DATE Civic TOTAL MAXIMUM @Max MINIMUM @Min FACTOR

More information

Monthly Trading Report Trading Date: Dec Monthly Trading Report December 2017

Monthly Trading Report Trading Date: Dec Monthly Trading Report December 2017 Trading Date: Dec 7 Monthly Trading Report December 7 Trading Date: Dec 7 Figure : December 7 (% change over previous month) % Major Market Indicators 5 4 Figure : Summary of Trading Data USEP () Daily

More information

ACCA Interactive Timetable

ACCA Interactive Timetable ACCA Interactive Timetable 2018 Professional information last updated uary 2018 v2.1 Please note: Information and dates in this timetable are subject to change. How the 4 exam sittings can work for you

More information

Econ671 Factor Models: Principal Components

Econ671 Factor Models: Principal Components Econ671 Factor Models: Principal Components Jun YU April 8, 2016 Jun YU () Econ671 Factor Models: Principal Components April 8, 2016 1 / 59 Factor Models: Principal Components Learning Objectives 1. Show

More information

Corn Basis Information By Tennessee Crop Reporting District

Corn Basis Information By Tennessee Crop Reporting District UT EXTENSION THE UNIVERSITY OF TENNESSEE INSTITUTE OF AGRICULTURE AE 05-13 Corn Basis Information By Tennessee Crop Reporting District 1994-2003 Delton C. Gerloff, Professor The University of Tennessee

More information

Supplementary Questions for HP Chapter Sketch graphs of the following functions: (a) z = y x. (b) z = ye x. (c) z = x 2 y 2

Supplementary Questions for HP Chapter Sketch graphs of the following functions: (a) z = y x. (b) z = ye x. (c) z = x 2 y 2 Supplementary Questions for HP Chapter 17 1. Sketch graphs of the following functions: (a) z = y x (b) z = ye x (c) z = x y. Find the domain and range of the given functions: (a) f(x, y) = 36 9x 4y (b)

More information

THE APPLICATION OF GREY SYSTEM THEORY TO EXCHANGE RATE PREDICTION IN THE POST-CRISIS ERA

THE APPLICATION OF GREY SYSTEM THEORY TO EXCHANGE RATE PREDICTION IN THE POST-CRISIS ERA International Journal of Innovative Management, Information & Production ISME Internationalc20 ISSN 285-5439 Volume 2, Number 2, December 20 PP. 83-89 THE APPLICATION OF GREY SYSTEM THEORY TO EXCHANGE

More information

EXAMINATIONS OF THE ROYAL STATISTICAL SOCIETY

EXAMINATIONS OF THE ROYAL STATISTICAL SOCIETY EXAMINATIONS OF THE ROYAL STATISTICAL SOCIETY GRADUATE DIPLOMA, 011 MODULE 3 : Stochastic processes and time series Time allowed: Three Hours Candidates should answer FIVE questions. All questions carry

More information

ACCA Interactive Timetable

ACCA Interactive Timetable ACCA Interactive Timetable 2018 Professional information last updated uary 2018 v4 Please note: Information and dates in this timetable are subject to change. How the 4 exam sittings can work for you ACCA

More information

A look into the factor model black box Publication lags and the role of hard and soft data in forecasting GDP

A look into the factor model black box Publication lags and the role of hard and soft data in forecasting GDP A look into the factor model black box Publication lags and the role of hard and soft data in forecasting GDP Marta Bańbura and Gerhard Rünstler Directorate General Research European Central Bank November

More information

HW #6. 1. Inflaton. (a) slow-roll regime. HW6.nb 1

HW #6. 1. Inflaton. (a) slow-roll regime. HW6.nb 1 HW6.nb HW #6. Inflaton (a) slow-roll regime In the slow-roll regime, we neglect the kinetic energy as well as f ÿÿ term in the equation of motion. Then H = ÅÅÅ 8 p 3 G N ÅÅÅ m f, 3 H f ÿ + m f = 0. We

More information

Four Basic Steps for Creating an Effective Demand Forecasting Process

Four Basic Steps for Creating an Effective Demand Forecasting Process Four Basic Steps for Creating an Effective Demand Forecasting Process Presented by Eric Stellwagen President & Cofounder Business Forecast Systems, Inc. estellwagen@forecastpro.com Business Forecast Systems,

More information

peak half-hourly New South Wales

peak half-hourly New South Wales Forecasting long-term peak half-hourly electricity demand for New South Wales Dr Shu Fan B.S., M.S., Ph.D. Professor Rob J Hyndman B.Sc. (Hons), Ph.D., A.Stat. Business & Economic Forecasting Unit Report

More information

Statistics for IT Managers

Statistics for IT Managers Statistics for IT Managers 95-796, Fall 2012 Module 2: Hypothesis Testing and Statistical Inference (5 lectures) Reading: Statistics for Business and Economics, Ch. 5-7 Confidence intervals Given the sample

More information

S95 INCOME-TESTED ASSISTANCE RECONCILIATION WORKSHEET (V3.1MF)

S95 INCOME-TESTED ASSISTANCE RECONCILIATION WORKSHEET (V3.1MF) Welcome! Here's your reconciliation Quick-Start. Please read all five steps before you get started. 1 2 3 Excel 2003? Are you using software other than Microsoft Excel 2003? Say what? Here are the concepts

More information

Least Squares and QP Optimization

Least Squares and QP Optimization Least Squares and QP Optimization The following was implemented in Maple by Marcus Davidsson (2012) davidsson_marcus@hotmail.com Stephen Boyd http://academicearth.org/lectures/least-squares Assume

More information

Optimal Investment Strategies: A Constrained Optimization Approach

Optimal Investment Strategies: A Constrained Optimization Approach Optimal Investment Strategies: A Constrained Optimization Approach Janet L Waldrop Mississippi State University jlc3@ramsstateedu Faculty Advisor: Michael Pearson Pearson@mathmsstateedu Contents Introduction

More information

ACCA Interactive Timetable

ACCA Interactive Timetable ACCA Interactive Timetable 2018 Professional Version 5.1 Information last updated 2nd May 2018 Please note: Information and dates in this timetable are subject to change. A better way of learning that

More information

Descriptive Statistics-I. Dr Mahmoud Alhussami

Descriptive Statistics-I. Dr Mahmoud Alhussami Descriptive Statistics-I Dr Mahmoud Alhussami Biostatistics What is the biostatistics? A branch of applied math. that deals with collecting, organizing and interpreting data using well-defined procedures.

More information

REPORT ON LABOUR FORECASTING FOR CONSTRUCTION

REPORT ON LABOUR FORECASTING FOR CONSTRUCTION REPORT ON LABOUR FORECASTING FOR CONSTRUCTION For: Project: XYZ Local Authority New Sample Project Contact us: Construction Skills & Whole Life Consultants Limited Dundee University Incubator James Lindsay

More information

How to find Sun's GHA using TABLE How to find Sun's Declination using TABLE 4...4

How to find Sun's GHA using TABLE How to find Sun's Declination using TABLE 4...4 1 of 8 How to use- TABLE 4. - GHA and Declination of the Sun for the Years 2001 to 2036- Argument Orbit Time How to find Sun's GHA using TABLE 4... 2 How to find Sun's Declination using TABLE 4...4 Before

More information

Marcel Dettling. Applied Time Series Analysis FS 2012 Week 01. ETH Zürich, February 20, Institute for Data Analysis and Process Design

Marcel Dettling. Applied Time Series Analysis FS 2012 Week 01. ETH Zürich, February 20, Institute for Data Analysis and Process Design Marcel Dettling Institute for Data Analysis and Process Design Zurich University of Applied Sciences marcel.dettling@zhaw.ch http://stat.ethz.ch/~dettling ETH Zürich, February 20, 2012 1 Your Lecturer

More information

ACCA Interactive Timetable

ACCA Interactive Timetable ACCA Interactive Timetable 2018 Professional information last updated uary 2018 v4.1 Please note: Information and dates in this timetable are subject to change. How the 4 exam sittings can work for you

More information

Average 175, , , , , , ,046 YTD Total 1,098,649 1,509,593 1,868,795 1,418, ,169 1,977,225 2,065,321

Average 175, , , , , , ,046 YTD Total 1,098,649 1,509,593 1,868,795 1,418, ,169 1,977,225 2,065,321 AGRICULTURE 01-Agriculture JUL 2,944-4,465 1,783-146 102 AUG 2,753 6,497 5,321 1,233 1,678 744 1,469 SEP - 4,274 4,183 1,596 - - 238 OCT 2,694 - - 1,032 340-276 NOV 1,979-5,822 637 3,221 1,923 1,532 DEC

More information

Average 175, , , , , , ,940 YTD Total 944,460 1,284,944 1,635,177 1,183, ,954 1,744,134 1,565,640

Average 175, , , , , , ,940 YTD Total 944,460 1,284,944 1,635,177 1,183, ,954 1,744,134 1,565,640 AGRICULTURE 01-Agriculture JUL 2,944-4,465 1,783-146 102 AUG 2,753 6,497 5,321 1,233 1,678 744 1,469 SEP - 4,274 4,183 1,596 - - 238 OCT 2,694 - - 1,032 340-276 NOV 1,979-5,822 637 3,221 1,923 1,532 DEC

More information

R = µ + Bf Arbitrage Pricing Model, APM

R = 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 information

SYSTEM BRIEF DAILY SUMMARY

SYSTEM BRIEF DAILY SUMMARY SYSTEM BRIEF DAILY SUMMARY * ANNUAL MaxTemp NEL (MWH) Hr Ending Hr Ending LOAD (PEAK HOURS 7:00 AM TO 10:00 PM MON-SAT) ENERGY (MWH) INCREMENTAL COST DAY DATE Civic TOTAL MAXIMUM @Max MINIMUM @Min FACTOR

More information

Monthly Trading Report July 2018

Monthly Trading Report July 2018 Monthly Trading Report July 218 Figure 1: July 218 (% change over previous month) % Major Market Indicators 2 2 4 USEP Forecasted Demand CCGT/Cogen/Trigen Supply ST Supply Figure 2: Summary of Trading

More information

2017 Settlement Calendar for ASX Cash Market Products ASX SETTLEMENT

2017 Settlement Calendar for ASX Cash Market Products ASX SETTLEMENT 2017 Settlement Calendar for ASX Cash Market Products ASX SETTLEMENT Settlement Calendar for ASX Cash Market Products 1 ASX Settlement Pty Limited (ASX Settlement) operates a trade date plus two Business

More information

Shape of the return probability density function and extreme value statistics

Shape of the return probability density function and extreme value statistics Shape of the return probability density function and extreme value statistics 13/09/03 Int. Workshop on Risk and Regulation, Budapest Overview I aim to elucidate a relation between one field of research

More information

Investigating Factors that Influence Climate

Investigating Factors that Influence Climate Investigating Factors that Influence Climate Description In this lesson* students investigate the climate of a particular latitude and longitude in North America by collecting real data from My NASA Data

More information