Search results

  1. A

    Spatial Econometrics - Bias?

    Dear all, I seem to be lost with the following problem in spatial econometrics: Assume the true model is a spatial error model (SEM). It is well known that using OLS for estimation is inefficient but still unbiased. Now, when I compare my OLS vs. SEM estimates in R, say, they are quite...
  2. A

    Added-Variable Plot for Panel Data

    Dear all, in Frees (2004): "Longitudinal and Panel Data" there is a short description of a procedure for Added-Variable or partial regression plots for Fixed Effects models: 1) Run Y on X leaving out x_j 2) Run x_j on "the rest" of X 3) Plot the respective residuals. My question: When I...
  3. A

    Panel Data: Vary in time but constant for all individuals

    Dear all, we are using a standard panel model like Y_it = a_i + b*X_it + e_it. For one additional variable (prices=P) we only have observations through time, but not across individuals, i.e. prices are the same for all individuals (think of "i" being the districts and prices are given for...
  4. A

    Gamma Poisson Mixture with finite Gamma

    Dear all I am working with a Gamma-Poisson mixture distribution where (and this is not usual) the support of the Gamma distribution (in fact, I am able to restrict to the neg exponential instead of the Gamma...) is finite, e.g. [0,1]. I would like to derive the mean and a Likelihood...
  5. A

    GARCH fitting to binary data / latent data

    Dear all I am trying to fit a simple ARCH(1)/GARCH(1,1) model to a set of binary data, i.e. I assume a latent GARCH process that is only observed at the values a and b, say (whenever it crosses or hits those thresholds). I found some ideas on fitting a censored GARCH (by SX Wei, for example...