Salinas Savings and Loan - a Forecasting Case

Salinas Savings and Loan is proud of its long tradition in Topeka, Kansas. Begun by Teresita Salinas 13 years
after World War II, the S&L has bucked the trend of financial and liquidity problems that has plagued the
industry since 1985. Deposits have increaseed slowly but surely over theyears, despite recessions in 1960,
1983, 1988, 1991, and 2001. Ms Salinas believes it is necessary to have a long-range strategic plan for her f
irm including a 1-year forecast and preferably even a 5-year forecast of deposits. She examines the past
deposit data and also peruses Kansas's Gross State Product (GSP), over the same 44 years. (GSP is
analogous to Gross Natinal Product, GNP, but on the state level.) The resulting data are in the following table:
YEAR
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980

DEPOSITS a
0.25
0.24
0.24
0.26
0.25
0.30
0.31
0.32
0.24
0.26
0.25
0.33
0.50
0.95
1.70
2.30
2.80
2.80
2.70
3.90
4.90
5.30

GSP b
0.40
0.40
0.50
0.70
0.90
1.00
1.40
1.70
1.30
1.20
1.10
0.90
1.20
1.20
1.20
1.60
1.50
1.60
1.70
1.90
1.90
2.30

YEAR
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
a
b

DEPOSITS a
6.20
4.10
4.50
6.10
7.70
10.10
15.20
18.10
24.10
25.60
30.30
36.00
31.10
31.70
38.50
47.90
49.10
55.80
70.10
70.90
79.10
94.00

GSP b
2.50
2.80
2.90
3.40
3.80
4.10
4.00
4.00
3.90
3.80
3.80
3.70
4.10
4.10
4.00
4.50
4.60
4.50
4.60
4.60
4.70
5.00

in$ millions
in$ billions

a ) Using nave approach, then moving average, with m = 3 and m = 5, then exponential smootihing with
a = .3, a = .5, and a = .7 (assume nave for starting forcast in 1960), then trend analysis, and finally
regression analysis, discuss which forecasting model fits best for Salina's strategic plan.
b ) Justify the selection of one model over another. (Tip: Obtain measures of forecast error - BIAS, MAD,
MSE and MAPE and construct a tabulated comparative summary). Where possible, compute
the coefficients of correlation and determination and interpret the results.