>plot apply*sumnews/stick=out smooth=lowess short

>let summi60 = sumnews - 60
>let sumge60 = 0
>if sumnews >= 60 then let sumge60 = 1
>format 3
>regress
>model apply = constant + sumnews + summi60*sumge60
>estimate

Dep Var: APPLY   N: 40   Multiple R: 0.781   Squared multiple R: 0.610
Adjusted squared multiple R: 0.589   Standard error of estimate: 36043.834

Effect         Coefficient    Std Error     Std Coef Tolerance     t   P(2 Tail)
CONSTANT        189663.521    11168.650        0.000      .      16.982    0.000
SUMNEWS           2187.969      329.938        1.319     0.266    6.631    0.000
SUMMI60
*SUMGE60         -2278.565      607.717       -0.746     0.266   -3.749    0.001

Analysis of Variance

Source             Sum-of-Squares   df  Mean-Square     F-ratio       P
Regression           7.53208E+10     2  3.76604E+10      28.988       0.000
Residual             4.80688E+10    37  1.29916E+09
-------------------------------------------------------------------------------
*** WARNING ***
Case           38 has large leverage   (Leverage =        0.568)

Durbin-Watson D Statistic          1.205
First Order Autocorrelation        0.376
 

>begin
>plot apply*sumnews/stick=out xmin=0 xmax=150 ymin=100000 ymax=400000
>fplot y = 189663.521 + 2187.969*x -2278.565*(x - 60)*(x >= 60);,
>stick=out xmin=0 xmax=150 ymin=100000 ymax=400000
>end