these two wards are time to time appeared in text book specially in research articles. Some explanation says that no difference between two concept of regression
Logistic regression, logit regression or logit model are the same; a regression model where the dependent variable (DV) is categorical. The binary logistic model is used to estimate the probability of a binary response based on one or more predictor (or independent) variables (features). Certain software, example like STATA have different commands to run this analysis, example, the logit and logistic commands. The main difference between the two is that the former displays the coefficients and the latter displays the odds ratios. You can also obtain the odds ratios by using the logit command with the 'or' option. So don't get confused, they are the same, it's just what you want to report!
The above answer is absolutely correct, there is no difference other than the terminology.
I would like to add that marginal effects are also a preferred tool, depending on the context of the research, to asses the impact of an explanatory variable.
In some programs (such as SPSS) logistic regression is the terminology used if all observations of the dependent variable are 0 or 1 and the dependent variables may be a combination of scale, ordinal or nominal variables. Another type of design would be to subject several experimental units to different levels of a treatment (e.g. a number of groups of 100 fruit flies being subjected to different dosages of an insecticide). The response variable is then the number (or %) of survivals. Thus X (the independent variable) is dosage and Y (the dependent variable) is % survival. In that case you may fit a logit model or a probit model, among others. In this case the crucial statistic is the LD50 (lethal dose at which the survival is 50%) which is a measure of the stength of the insecticide. The logistic regression model and the logit model have the same form but the input is different.
- Badges
- Science topic
- Similar topics
- Books