May 2016 - Gretl-users - gretlml.univpm.it

by Henrique Andrade

Dear Gretl Community, I really stuck trying to define a function that gives a power set of a set. Suppose I have a set S: S = {"A", "B", "C"} The associated power set, P(S), is: P(S) = {{ }, {"A"}, {"B"}, {"C"}, {"A", "B"}, {"A", "C"}, {"B", "C"}, {"A", "B", "C"}} All that I can think by now (shame on me!) is this: strings S = defarray("A", "B", "C") scalar P_S_len = 2^nelem(S) # the size of the power set strings P_S = array(P_S_len) # an array with 8 spaces. Does anyone have any ideas? Best, Henrique Andrade

6 years, 2 months

5
15
0 / 0

warning: gdt-reading bug

by Allin Cottrell

We've just noticed that a bug was introduced into our code for reading native gretl .gdt data files in August of this year. The bug should be triggered only rarely, but we thought it wise to issue a warning. Description of bug: If a gdt file contains "subnormal" values (that is, floating point values that are too close to zero to be represented with the usual precision), then when such a file is read on Linux, the first subnormal value to be found on a given row (observation) will be incorrectly copied into the remaining columns (series) on that row. Example: A gdt file containing 10 series has a subnormal for series number 5 on row 25. Then when the file is read on Linux, that subnormal will replace the correct values for series 6 to 10 for observation 25. Comment: This won't affect the reading of "primary" data (actual micro- or macroeconomic measurements), which will never contain subnormal values (we're talking about absolute values less than 10 to the minus 307). And the bug is not triggered on MS Windows. However, subnormal values may be produced by some data transformations (such as squaring very small numbers, or computing the normal CDF of very big negative values). Fix: This is now fixed in the git source for gretl and also the current snapshots. And we will put out a new release soon, gretl 2015d. Diagnostic: If you think a dataset may suffer from this problem, you can run the script checkdata.inp, from http://ricardo.ecn.wfu.edu/pub/gretl/checkdata.inp First load the dataset in question. Then open checkdata.inp and run it. An affected dataset may produce something like this: <script-output> Total number of values examined: 164122 Check for subnormal floating-point values ----------------------------------------- Total number found: 138 Longest (row) sequence: 138 (occurs at obs 210, starting series ID 461) Number of sequences (of length >= 2): 1 </script-output> The symptom of a problem is that we find a consecutive sequence of subnormal values on one or more rows of the dataset. This could occur for "natural" reasons but it may indicate corruption. Isolated subnormals don't indicate the bug. And again, most datasets should contain no subnormal values. Allin Cottrell

8 years, 2 months

3
7
0 / 0

plots with "bands"

by Allin Cottrell

This is somewhat relevant to http://lists.wfu.edu/pipermail/gretl-users/2016-May/011845.html but may be of interest in its own right. In git and snapshots there's a new option, --band, for the commands "gnuplot" and "plot". Right now it's very basic (and not yet documented); it could be fancied up in future by adding a parameter to inflect the option's effects. Anyway, here's how it works now. If you have a plot with 3 or more "lines", the --band option tells gretl to interpret lines 2 and 3 as upper and lower limits (or vice versa) of a band of some sort (e.g. a confidence interval). This means that 1) The two band limits will be shown in the same color. 2) The "key" or legend will be suppressed. 3) Gretl will not add a second y-axis, regardless of the scale of the data (i.e. the option --single-yaxis is implicit). To see what this does in practice you can compare these two plots: <hansl> nulldata 50 set seed 12345 setobs 1 1 --special-time series y = normal() series upper = y + 2 series lower = y - 2 list L = y upper lower gnuplot L --time-series --with-lines --output=display gnuplot L --time-series --with-lines --output=display --band </hansl> Yes, it would be nice if you could also set the style of the band (e.g. make it a shaded area rather than two lines). That may come, but it's a lot more complicated. Allin

8 years, 6 months

4
5
0 / 0

odd behavior of "diff"

by Summers, Peter

Dear all, I've run across something strange when using diff() in a loop. Given a series with missing values, Gretl seems to be "recycling" values from the previous loop iteration. The following script illustrates the problem: on the second loop iteration, the series "r" should be missing for the first half of the sample but instead the values from the first loop iteration are printed. Doing the same thing "by hand" after the loop produces what I'd expect. Is this intended? <hansl> nulldata 104 setobs 52 2006-01-03 series y1 = uniform() series y2 = uniform()*(obs>obsnum(2006-07-04)) y2 = zeromiss(y2) # y2 is missing for first half list ylist = y1 y2 loop foreach i ylist --quiet r = 100*diff(log(ylist[i])) print ylist[i] r --byobs endloop r1 = 100*diff(log(y1)) r2 = 100*diff(log(y2)) print r1 r2 --byobs </hansl> TIA, PS [cid:logoimg]

8 years, 6 months

4
5
0 / 0

removing nan and inf from a matrix

by Logan Kelly

Hello, I need to take the log difference of a matrix, i.e. log(M[2 rows(M):,]/M[1:rows(M)-1,]). Unfortunately, M has elements equal to zero. I need to replace the nan's and inf's with 0's. This almost works M = isnan(M) ? 0 : M but does not remove inf's. Any sugestions?

8 years, 6 months

5
8
0 / 0

Double loop

by Ma. jo Arteaga Garavito

Hi! I’m new to gretl programming. Please, can anyone tell me why my code doesn’t recognize my double loop. I don’t understand. # a 14 de mayo de 2016 # función que genera la contribución de Shapley include drop_majo.inp include R2_majo.inp function matrix shapley(matrix X "Variables explicativas", matrix y "Variable dependiente") scalar T=rows(X) #número de observaciones scalar N=cols(X) #número de variables explicativas scalar M=2^N #número de regresiones factibles scalar M2=M-1 scalar N2=N-1 matrix Z = zeros(M,1) matrix C = zeros(1,N) loop k=0..M2 matrix Xs=X loop for (i=N; i=1;i-=1) if floor((k%(2^i))/2^(i-1))==0 matrix Xs=dropcol(Xs,i) endif endloop matrix Z[k+1]=R2(Xs,y) endloop print Z loop j=1..N loop m=0..M2 if floor(m%(2^j)/2^(j-1))==0 matrix C[j]=C[j]+Z[m+1] else matrix C[j]=C[j]-Z[m+1] endif endloop matrix C[j]=C[j]/(2^(N-1)) endloop print C end Also, sometimes the console freezes. It doesn’t respond to anything I write, so I have to constantly close Gretl. Thanks! María José Enviado desde Correo para Windows 10

8 years, 7 months

3
3
0 / 0

Holt-Winters package

by Raul Gimeno

Hello I've been using the Holt-Winters package but I cannot replicate my Excel-calculation results with this package. The starting value from the package for the trend is 245 mine is 166.396. By running a regression on the full sample I get completely different results for these starting values, although the same methodology as described in the help description has been used. For replication purposes I send my excel spreadsheet and I would be glad to understand how these starting values have been effectively calculated. Thank you for your help Raul Gimeno **

8 years, 7 months

4
4
0 / 0

Re: [Gretl-users] Gretl's current Random Number Generator?

by Alecos Papadopoulos

Good afternoon. Has anything changed in Gretl's RNG after Yalta & Schreiber's (2012), "Random Number Generation in gretl", Journal of Statistical Software ? I need to put the information in a paper (not any detailed technical description, just a sentence or two). Thank you. -- Alecos Papadopoulos PhD Candidate Athens University of Economics and Business, Greece School of Economic Sciences Department of Economics https://alecospapadopoulos.wordpress.com/

8 years, 7 months

2
1
0 / 0

Fractional Differencing / Integration

by Fernando Fernandes Neto

Hello folks. I'd like to know if there is a simple way of making a fractional integration of a time series. I.e., I've built a model for a fractional differenced series. Then, I'd like to obtain the original level series again, for comparison purposes with other traditional models. How can I do that? Thanks in advance, Fernando Fernandes Neto

8 years, 7 months

3
3
0 / 0

pooled probit - normality test

by Jan Tille

Dear list members, I have a question regarding the normality test that is printed out if one estimates a pooled probit model. The user guide says it is a conditional moment test on skewness and kurtosis (p.285). The next paragraph begins with "In this context $uhat...." and references to Gourieroux et al. (1987). I understand that the normality test is conducted using the generalized residuals. Is this test also presented in their paper? If not, would it be possible to provide a reference? Thanks in advance and best regards, Jan

8 years, 7 months

2
2
0 / 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Gretl-users May 2016