The intercept sometimes appears in model dissection reports. What can the intercept's weight tell me about my model?
This email and any files included with it may contain privileged,
proprietary and/or confidential information that is for the sole use
of the intended recipient(s). Any disclosure, copying, distribution,
posting, or use of the information contained in or attached to this
email is prohibited unless permitted by the sender. If you have
received this email in error, please immediately notify the sender
via return email, telephone, or fax and destroy this original transmission
and its included files without reading or saving it in any manner.
It tells you the base rate of positive cases. The scale is log-odds and if
you have any very common features then those should be included in this
estimate. L_1 regularization will tend to prefer the intercept over any
feature that has less than 100% prevalence. If you have features with 100%
prevalence and constant weight then you might as well eliminate them anyway.
To convert a log-odds value x to a probability, use 1/(1+exp(-x))
On Wed, Dec 15, 2010 at 12:06 PM, Adrian E. Gould <[hidden email]> wrote:
> The intercept sometimes appears in model dissection reports. What can the
> intercept's weight tell me about my model?