Data Analysis and Statistics (Mathematics)

Mathematics

残差の解析

適合の"度合い"の尺度の一つに残差があります。これは、観測データと予測データの差で表すものです。正規化したcdate 値を使って、種々の適合に対する残差の比較を行ないましょう。

Fit
Residuals

p1 = polyfit(sdate,pop,1);
pop1 = polyval(p1,sdate);
plot(cdate,pop1,'-',cdate,pop,'+')

res1 = pop - pop1;
figure, plot(cdate,res1,'+')

p = polyfit(sdate,pop,2);
pop2 = polyval(p,sdate);
plot(cdate,pop2,'-',cdate,pop,'+')

res2 = pop - pop2;
figure, plot(cdate,res2,'+')

p = polyfit(sdate,pop,4);
pop4 = polyval(p,sdate);
plot(cdate,pop4,'-',cdate,pop,'+')

res4 = pop - pop4;
figure, plot(cdate,res4,'+')

Fit	Residuals
`p1 = polyfit(sdate,pop,1);` `pop1 = polyval(p1,sdate);` `plot(cdate,pop1,'-',cdate,pop,'+')`	`res1 = pop - pop1;` `figure, plot(cdate,res1,'+')`
`p = polyfit(sdate,pop,2);` `pop2 = polyval(p,sdate);` `plot(cdate,pop2,'-',cdate,pop,'+')`	`res2 = pop - pop2;` `figure, plot(cdate,res2,'+')`
`p = polyfit(sdate,pop,4);` `pop4 = polyval(p,sdate);` `plot(cdate,pop4,'-',cdate,pop,'+')`	`res4 = pop - pop4;` `figure, plot(cdate,res4,'+')`

多項式近似指数近似