# dummy variable multiple regression

#### Manasa

##### New Member
Hello guys,

I am doing a project on calendar affects(seasonality) in the stock market for which i am trying to perform a dummy variable regression using excel and also minitab and i think i have landed in a dummy variable trap. Have tried reading a lot about it but could not solve the problem.

My dependent variable is the mean monthly returns from the year 2000 - 2010. Have 11 independent variables for 11 months.

Y = constant + beta1{Dummy(feb)} + beta2{Dummy(march)} + beta3{Dummy(april)} + beta4{Dummy(may)} + beta5{Dummy(june)} + beta6{Dummy(july)} + beta7{Dummy(aug)} + beta8{Dummy(sept)} + beta9{Dummy(oct)} + beta10{Dummy(nov)} + beta11{Dummy(dec)}

where Y = Monthly returns of Nifty
Dummy(feb)= 1 if the month is Feb; 0 otherwise
Dummy(March) = 1 if the month is March; 0 otherwise
Dummy(april) = 1 if the month is April; 0 otherwise
and so on.

Have taken january as the benchmark month and hence excluded the dummy variable for month of jan. i.e for the month of jan, have all 0s as independent variables.

The result of regression analysis shows t-stat values as #NUM!, standard error as 0. I request someone to please help me out.

#### duskstar

##### New Member
Hi,

Just out of curisoity, why haven't you just included months as a cateogrical variable with January as the baseline?