PsychProbability - aleslab/Psychtoolbox-3-aleslab-fork GitHub Wiki

Psychtoolbox:PsychProbability.

Computation/Evaluation of probability distributions, basic sampling and
randomization.

BalanceFactors - Balance a set of factors given the factor levels.
BalanceTrials - Balance a set of factors given the factor levels. Takes alternative parameters wrt. BalanceFactors.
BuildMarkovK - Build covariance matrix for Markov process.
ChiSqrCumulative - Chi-squared distribution probability function.
ChiSquarePDF - Chi-squared distribution PDF.
ChooseKFromN - Randomly choose k distinct integers out of n.
ClockRandSeed - Seed rand() and randn() from clock.
CoinFlip - Bernoulli random variable.
CoinFlipPDF - Bernoulli distribution PDF.
ColMean - Take column means, works for one row matrices.
ColStd - Take column stds, works for one row matrices.
CombVec - Generate all possible combinations of input vectors.
CovToCorr - Compute matrix of correlations from covariance.
FindNeighborCorr - Find nearest neighbor correlations in data set.
MultiNormalDraw - Draw vectors from multivariate normal.
MultiNormalPDF - Compute multivariate normal PDF.
NormalCumulative - Univariate normal CDF
NormalDraw - Draw from univariate normal.
NormalPDF - Univariate normal PDF.
NormalProb - Compute probability of univariate normal in interval.
NRandPerm - Randomly select elements from a random permutation.
RandDim - Randomize matrix along any dimension.
Randi - Get random integer sample. Same as Ranint, but Denis prefers it.
RandLim - Generates a matrix of random numbers between a lower and upper limit.
RandSample - Get a random sample from a list.
RandSel - Randomly select elements from matrix, elements are replenished.
Ranint - Get random integer sample. Same as Randi, but David prefers it.
Sample - Get a random sample from a list.
Shuffle - Randomly reorder the entries of vector/matrix.
UniqueAndN - Returns unique elements of array and the number of times each element occurs in the array.
UniqueNoSorting - Returns elements of array in order of appearance or disappearance.
URandSel - Randomly select elements from matrix, elements are not replenished.

Combining CombVec and RandDim
Using CombVec and RandDim, it is very simple to generate all trials
in an experiment with a factorial design and then randomize these
trials:
cond_a = [1 2]; cond_b = [120 180]; ntrial = 2;
trial = CombVec(cond_a,cond_b);
trial =
1 2 1 2
120 120 180 180

trial = repmat(trial,1,ntrial); % each combination of conditions occurs ntrial times
trial(end+1,:) = [1:size(trial,2)]; % append trial number
trial = RandDim(trial,2); % shuffle trial order

trial =
2 1 1 2 1 1 2 2
120 180 180 180 120 120 120 180
6 3 7 8 5 1 2 4

BalanceFactors.m was contributed by David E. Fencsik ([email protected]), thanks!

{{category}}