Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, is already up and running.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Re: Regressing the Habit Persistence Model

From   Anders Alexandersson <>
Subject   Re: st: Re: Regressing the Habit Persistence Model
Date   Fri, 20 Aug 2010 10:03:40 -0400

Nabil, no -regress- is not suitable for serially correlated data.
Do you need serially lagged dependent variable(s) or are alternative
methods okay?
The references you provided referred to several habit persistence
models, which means I am still confused about what you want.
I suggest the time series command -prais- with the default option
rhotype(regress) which uses GLS rather than lagged dependent
If you need lagged dependent variables, consider -xtdpdsys-.

Anders Alexandersson

On Thu, Aug 19, 2010 at 4:57 PM, Nabil Hashmi
<> wrote:
> The following two links have more information about what I mean by the
> habit persistence model. Consumption in one time period is partially
> determined by consumption in the time period that came immediately
> before it, i.e. I want to regress at least one lagged dependent
> variable. My best guess is that this would be done through the
> 'regress' command, although I am not sure if there is another, more
> suitable command. If using the regress command, I would like to find a
> way to include information about the coefficients and standard errors
> of rho and sigma-squared.
> Thanks!
> On Thu, Aug 19, 2010 at 4:48 PM, Anders Alexandersson
> <> wrote:
>> Nabil, please provide a reference and/or a description of the "habit
>> persistence" model.
>> Are you referring to the -regress- command or to some other Stata
>> regression command?
>> Anders Alexandersson
>> On Thu, Aug 19, 2010 at 4:19 PM, Nabil Hashmi
>> <> wrote:
>>> I am having difficulty regressing a habit persistence model in Stata,
>>> but I am having some difficulties. I am trying to replicate results
>>> from another study, so I would also like to have the rho and
>>> sigma-squared values included in the output of the regression so I can
>>> see if my results do replicate the initial findings. What is the best
>>> way to regress a habit persistence model in Stata? Thanks!
>>> --
>>> Nabil Hashmi
>>> Principal of Finance, Compass Partners |
>>> Georgetown University | Walsh School of Foreign Service | Class of
>>> 2011 | International Political Economy
>>> (860) 944-7379
>> *
>> *   For searches and help try:
>> *
>> *
>> *

*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index