This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: A d v a n c e d P r o g r a m m i n g i n Q u a n t i t a t i v e E c o n o m i c s Introduction, structure, and advanced programming techniques 17 – 21 August 2009, Aarhus, Denmark Charles Bos [email protected] VU University Amsterdam Tinbergen Institute Advanced Programming in Quantitative Economics – p. 1 Day 4  Morning 9.00L Topics ◦ Finish standard errors, see Day 3b ◦ Including packages ◦ Including magic numbers ◦ Including graphs 10.30P Estimating a duration model ◦ Transform . 5 &lt; β 2 &lt; 1 ◦ Graph the durations ◦ Advanced: • Draw N = 1000 ,y i ∼ N (0 ,σ 2 ) for a σ of choice. Make a QQ plot using DrawQQ • Make the QQ plot ‘by hand’ using DrawXMatrix , drawing the empirical quantiles of the y ’s against the theoretical quantiles of the normal density • Make a residual plot E i = (Λ i y i ) α for your y ’s of the duration model, and a QQplot against the Exp(1) density Advanced Programming in Quantitative Economics – p. 2 Day 4  Afternoon 12.00 Lunch 14.30 Seminar Sir David Hendry, one of the founding fathers of OxMetrics... 16.00P Let’s call it tutorial, for those who want to finish something 17.00 The end for today Advanced Programming in Quantitative Economics – p. 3 Include Enlarging the capabilities of ox beyond oxstd.h capabilities: Either #include &lt;oxprob.h&gt; (to include the mentioned file literally within the program at that point, and will be compiled in), or #import &lt;maximize&gt; (to import the code when needed; precompiled code is used when available) Advanced Programming in Quantitative Economics – p. 4 Oxprovided packages Package Purpose oxprob.h Extra probability densities oxfloat.h Definition of constants oxdraw.h Graphics capabilities ( * ) arma.h ARMA filters and generators quadpack.h Univeriate numerical integration maximize Optimization using GaussNewton or Simplex methods ( * ) maxsqp Maximize nonlinear function with sequential quadratic programming solvenle Solve a system of nonlinear equations solveqp Solve a quadratic program with restrictions database General class for creating a database modelbase General class for building a model simulation General class for simulation exercise Advanced Programming in Quantitative Economics – p. 5 Userprovided packages Package Author Purpose arfima Doornik, Ooms Long memory modelling dcm Eklof, Weeks Discrete choice models dpd Doornik, Arellano, Bond Dynamic Panel Data models financialnr Ødegaard Financial numerical recipes gnudraw.h Bos Alternative graphing capabilities maxsa.h Bos Simulated Annealing mc2pack.h Bos Markov chain Monte Carlo estimation msvar Krolzig Markov switching oxutils.h Bos Some convenient utilities ( * ) oxdbi Bruche A database independent abstraction layer for Ox ssfpack.h Koopman, Shephard, Doornik State space models ... ...and many others [email protected] Laurent, Urbain Use CML optimisation in OxGauss oxgauss Doornik Run Gauss code through Ox • Packages reside either in oxhome/packages , or in a local packages folder.packages folder....
View
Full Document
 Spring '04
 KimCJ
 Economics, Econometrics, Computer Programming, Optimization, Quadratic programming, Quantitative economics

Click to edit the document details