Suppose, one of my independent variables is index number such as d= 1, 2, 3,......15, where higher value shows higher corruption(say) and lower value shows lower corruption. If the sign of coefficient is negative then how can I explain this coefficient?
If you treated the independent value as continuous variable, yes, that is the correct explanation to the sign.
If the sign is negative, it means that for an increase of the corruption, the dependent variable decreases as per the coefficient value.
Genc Alimehmeti si correct but you to take into account also the following:
1) if you have a linear regression model then the changes are marginal meaning 1 unit of corruption in plus ===> "coeficient" units in minus for the dependent variable and viceversa
2) if you have a log-log model then you have elasticities (procentual changes)
3) if you have a logit model with corruption as factor then you have a unit of change of corruption(+) ===> a unit of chnage of the logit(results) (-) and from here....and also might be other particularities....
Suppose, for another index 1= pure peg exchange rate regime, 2= ...., 3=......, 4=......
5= pure floating, then if growth rate = f( exchange rate regime), then what does it mean? Same as before? Bogdan-Vasile Ileanu
I do not suggest to use classic regression with a X variable having the values 1 to 5. You do not obey the restrictions of continuos and even normality of X (covariates=factor). What you can do is to use 5-1 dummy(1/0) covariates. Here the economic/practical significance of the coefficients is a little bit different
If your index variable is one of the many independent variables in the model, the explanation for that variable is same as for coefficients of the other variables. For example, a positive sign means for an increase in one unit of the index variable, the dependent variable increases by the amount of the coefficient. Conversely, a positive sign means for a increase in one unit of the index variable, the dependent variable decreases by the amount of the coefficient
Suppose, for another index, exchange rate regime: 1= pure peg exchange rate regime, 2= ...., 3=......, 4=...... 5= pure floating, then
If the function is growth rate = f( exchange rate regime, x1, x2, .....xn), then if we find the coefficient of exchange rate regime is negative, can we say,
(1) flexible regime is better for growth?
(2) flexible regime negatively affect the growth?
(3) which of the regime negatively affect the growth?
(4) none of them?
If those questions are incorrect, then in terms of regime index, what should we say about the relationship between exchange rate regime and growth rate?
Venkatarao Nagubadi
- Badges
- Science topic
- Similar topics
- Linguistics
- Composition Studies
- Publication