An exogenous variable in the context of regression analysis is a variable which is not affected by other variables. Once introduced into the system it is regarded as fixed. On the other hand a variable is said to be endogenous if its value is determined by other variables.
An exogenous variable in the context of regression analysis is a variable which is not affected by other variables. Once introduced into the system it is regarded as fixed. On the other hand a variable is said to be endogenous if its value is determined by other variables.
Endogenous are dependent variable (DV) that we want to explain. Exogenous are independent variables (IVs) that is/are effecting the dependent variable. The key to identify which is which, is that the endogenous variable has NO impact on the exogenous variables. Now, in multivariate regression analysis, commonly we want to estimate if there is statistical relationship between a DV and IVs. The analysis depends on type of DV. If the DV is a continuous variable and reasonably normal, we use linear regression. If dichotomous, we use logistic regression. If multinominal (or Poisson), we use log-linear analysis. Or Cox regression for time-to-event experiments (survival). As you notice, the terms endogenous and exogenous are not commonly used when explaining regression analysis in general. For DV, we usually use the term such as a response, an outcome or criterion variable. For IVs, we usually use the term such as predictors, the causes, or explanatory variables. Although it looks like it falls into similar definition, it is uncommon (as far as I know in medical research that is) to use the term endogenous/exogenous. It is, however, commonly used in the structural model of Structual Equation Modeling (SEM). Hope this helps make it a bit clearer. Best wishes, Hanif.
Completely agree with Rahman! Furthermore, would like to add few information on the terms- endogenous and exogenous that are usually used in the Structural Equational Modeling (SEM) because an endogenous variable is one that is influenced by other factors (i.e. called exogenous variables) in the system. A model (i.e. in a system) should be followed through so as to run a SEM.
Especially in Structural equation modeling (SEM), , we need to make a distinction between the endogenous factors and exogenous factors. An exogenous variable is a variable which is not affected by other variables. On the other hand a variable is said to be endogenous if its value is determined by other variables.
But there are many Independent variables existing out of them some are related to the experiment and some are not connected to the experiment under consideration , but these IV's which are not belonging to the experiment but influence the DV and such IV's (Extrenous variables) need to be controlled.and there many methods for controlling the extranous variables.
Hope this will be help you to understand, what you needed. Otherwise you be specific what is your study so we can discuss more to bring more clarity.
I fully agree with all mentioned above sincerely. Moreover, I also would like to add some information:
-Exogenous means determined outside of the model which is latent variables (external sources);
-Endogenous means determined within the model which internal sources (latent variables);
Additionally: If something is exogenous (IV), we pretty much have to make up a value for it or do some research find a value for it and plug it into our model. For the most part, it's going to be fixed or not changing in our model.
For endogenous (DV), it's determined within the model we're using our model to find a value for these endogenous variables. As Mamunur Rashid mentioned above correctly that these two variables mostly used in SEM.