multivariate ordered logit regression
[res]=ologit(namey,arg1,...,argn)
* namey = a time series, a real (nx1) vector or a string equal to the name of a time series or a (nx1) real vector between quotes which should be coded from 0 to ncat
* argi = an argument that can be:
- a time series
- a real (nx1) vector
- a real (nxk) matrix
- 'init=v' where v is the optional vector of starting values
- the string 'noprint' if the user doesn't want to print the results of the regression
- the string 'dropna' if the user wants to use in the regression all dates with no NA value in any variable (the main use of this option should be when dealing with daily ts)
- 'optfunc=optim' if the user wants to use the optim optimisation function (default: optimg)
- 'opt_nelmead=crit,nitermax' with crit the value of the convergence criterion in the Nelder-Meade optimisation function and nitermax the maximum number of iterations (default = 'opt_nelmead=2*%eps,1000')
- 'opt_optim=opts' where opts are options for optim that can be entered after the starting value of the parameters (default = 'opt_optim=,''ar'',1e6,1e6'')
- 'opt_convg=val' where val is the threshold on gradient norm (default = 'opt_convg=1e-5')
* res = a results tlist with
- res('meth') = 'ordered logit'
- res('y') = (nobs x ncat) matrix of data
- res('x') = (nobs x nx) matrix of data
- res('nobs') = number of observations
- res('nvar') = number of variables
- res('ncat') = number of categories of dependent variable (including the reference category j = 0)
- res('beta') = (nvar*ncat x 1) vector of beta coefficients: [beta_1 ; beta_2 ; ... ; beta_ncat] under normalization beta_0 = 0
- res('yhat') = (nobs x ncat) matrix of fitted values probabilities: [P_0 P_1 ... P_(ncat-1)] where P_j = [P_1j ; P_2j ; ... ; P_nobsj]
- res('r2mf') = McFadden pseudo-R^2
- res('rsqr') = Estrella pseudo-R^2
- res('llike') = unrestricted log likelihood
- res('lratio') = LR test statistic against intercept-only model (all bets=0), distributed chi-squared with (nvar -1)*ncat degrees of freedom
- res('covb') = (nvar*ncat x nvar*ncat) covariance matrix of coefficients
- res('tstat') = (nvar*ncat x 1) vector of t-statistics
- res('pvalue') = (nvar*ncat x 1) vector of corresponding p-values
- res('namey') = name of the y variable
- res('namex') = name of the x variables
- res('prests') = boolean indicating the presence or absence of a time series in the regression
- res('dropna') = boolean indicating if NAs have been dropped
- res('bounds') = if there is a timeseries in the regression, the bounds of the regression
- res('nonna') = vector indicating position of non-NAs
load(GROCERDIR+'\data\gunther.dat') res=ologit('rrating',['asset' 'equ' 'growth' 'loa' 'metro' 'prl']) / Example taken from ologit_d, provides the ordered multivariate logit regression // of variable 'rrating' (ratings of credit institutions) // on variables 'ass' (logarithm of assets), 'equ' (equity to assets ratio), // growth (GDP growth rate of the state where the bank is operating), // 'loa' (loans to assets ratio), 'metro' (1 if metropolitan statistic area) // and 'prl' (no performing loans to total loans ratio), // all variables taken from a database provided by J. Gunther (Gunther J. (1999): // "Between a rock and a hard place: the CRA-safety and soundness pinch", // Federal Reserve Bank of Dallas, Economic and Financial Review, Second Quarter 1999). | ![]() | ![]() |