The accompanying dataset shows the monthly coal consumption (in trillions of BTUs) over the course of three years. Develop a multiple regression model with categorical variables that incorporate seasonality for forecasting coal consumption.

Month and Year Consumption Month an 1990 16.69 1 eb 1990 15.39 2 Mar 1990 12.96 3 Apr 1990 12.51 4 May 1990 8.55 5 un 1990 8.84 6 ul 1990 12.58 7 Aug 1990 11.74 8 Sep 1990 9.68 9 Oct 1990 9.78 10 Nov 1990 14.65 11 Dec 1990 24.71 12 an 1991 20.15 13 eb 1991 14.2 14 Mar 1991 12.71 15 Apr 1991 9.53 16 May 1991 7.9 17 un 1991 7.66 18 ul 1991 10.1 19 Aug 1991 9.21 20 Sep 1991 7.57 21 Oct 1991 8.39 22 Nov 1991 15.9 23 Dec 1991 20.38 24 an 1992 17.19 25 eb 1992 13.66 26 Mar 1992 12.36 27 Apr 1992 12.5 28 May 1992 7.63 29 un 1992 7.04 30 ul 1992 11.17 31 Aug 1992 9.3 32 Sep 1992 8.72 33 Oct 1992 8.75 34 Nov 1992 15.01 35 Dec 1992 21.4 36