liveBook · Manning

Squared error

As explained in Statistics Every Programmer Needs

The sum of squared residuals is the sum of the squared differences between observed response values and their fitted values.

Squared loss sums the squared residuals between observed response values and fitted values: SSE = Σ(y − ŷ)². Linear regression chooses coefficients that minimize this quantity, so smaller SSE indicates a closer fit to the observed data. section 4.1.2 section 4.2.3

For a fitted regression, squared loss is the unexplained portion of total variation. With SST as total variation, SSR as explained variation, and SSE as error, SST = SSR + SSE; R² = SSR / SST therefore reports the explained proportion. section 4.2.3

Computing SSE from a fitted model

For the marathon stage1-to-stage2 regression, calculate total variation and explained variation, then obtain squared loss either by subtraction or directly from residuals: SSE = SST − SSR SSE = 49,636.95 − 28,537.79 = 21,099.16 In Python: SSE = np.sum(lm.resid ** 2) The direct calculation returns approximately 21,099.16, matching the decomposition. The same model gives R² ≈ 0.575, meaning the regression explains about 57.5% of the response variation. section 4.2.3

Table 4.2 The sum of squares summary for the fitted linear regression. The stage1 and stage2 columns represent, respectively, the observed values from the two mds variables. SST equals the sum of squares total, or the total variability in the dependent variable without the consideration of any predictors. The yhat column contains the predictions for the response variable from the fitted regression. SSE equals the sum of squares error, or the portion of the SST that can’t be explained by the regression; and SSR equals the sum of squares regression, or the percentage of the variance in the dependent variable explained by the model.
x
stage1
y
stage2
( y − ȳ) 2
SST
y
yhat
( y − ŷ) 2
SSE
( ŷ − ȳ) 2
SSR
152
191
3,654.20
195.78
22.88
3098.73
162
184
4,549.50
208.87
618.36
1813.33
165
193
3,416.40
212.79
391.71
1494.46
163
191
3,654.20
210.18
367.68
1,703.62
166
193
3,416.40
214.10
445.21
1,395.02
187
292
1,644.30
241.57
2,542.75
97.53
175
235
270.60
225.87
83.27
654.10
202
226
647.70
261.20
1,238.96
95.04
179
214
1,402.50
231.11
292.68
413.80
222
274
508.50
287.36
178.62
1,289.87
184
300
2,357.10
237.65
3,887.60
190.46
193
242
89.30
249.42
55.12
4.10
255
279
759.00
330.54
2,656.24
6,255.02
202
288
1,335.90
261.20
718.30
95.04
221
269
308.00
286.06
290.92
1,197.61
207
329
6,014.00
267.74
3,752.75
265.37
227
268
273.90
293.91
671.14
1,802.54
239
315
4,038.60
309.61
29.10
3,382.11
164
201
2,545.20
211.48
109.90
1,597.33
226
345
8,751.60
292.60
2,745.97
1,693.16
49,636.95
21,099.16
28,538.24

Inspect the sum-of-squares table to compare observed responses, fitted values, residual squares, SST, SSE, and SSR row by row. section 4.2.3

How Statistics Every Programmer Needs covers it

How deep it goes
Derives SSE through residuals and SST–SSR decomposition, but stops short of discussing optimization algorithms or loss-function variants.
Examples
Uses marathon stage1 and stage2 times, fitted values, a sum-of-squares table, and NumPy calculations.
What it assumes
Builds from observed values, fitted values, means, residuals, and basic linear-regression coefficients.
What you can do after
Evaluate regression fit by calculating SSE directly from residuals and validating it against SST minus SSR.
How it connects
Connects squared loss to linear regression, residuals, goodness of fit, R-squared, SST, and SSR.

Also explained in

Compare how 5 books explain squared error →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Statistics Every Programmer Needs. The section links point to the parts of the book it draws on.