* Segmented Analysis of Russian Casualties in Ukraine on Fri Sep 11 16:44:29 2026 - input data directory: ./results - results directory: ./results - transcript to: ./results/segmented-ukr-rus-casualties-transcript.txt Loading updated data: --------------------- * Updated data file: ./results/russian-casualties-in-ukraine-updated-2026-09-11.tsv - Found 126 rows x 3 columns: - Columns: DayNum, Date, Soldiers Doing 3-fold crossvalidated segmented fits: ------------------------------------------- - Segmented fit, testFold = 1 o Train data: 84 points o Test data: 42 points o Simple linear fit, no kinks: Call: lm(formula = Soldiers ~ DayNum, data = trainData) Residuals: Min 1Q Median 3Q Max -98214 -8140 4030 11005 54730 Coefficients: Estimate Std. Error t value Pr(>|t|) (Intercept) 1.030e+05 2.378e+03 43.3 <2e-16 *** DayNum 1.013e+03 7.978e+00 127.0 <2e-16 *** --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 Residual standard error: 19310 on 82 degrees of freedom Multiple R-squared: 0.9949, Adjusted R-squared: 0.9949 F-statistic: 1.612e+04 on 1 and 82 DF, p-value: < 2.2e-16 o Davies test for need of kink: Davies' test for a change in the slope data: formula = Soldiers ~ DayNum , method = lm model = gaussian , link = identity segmented variable = DayNum 'best' at = 419.67, n.points = 10, p-value < 2.2e-16 alternative hypothesis: two.sided o Segmented fit: ***Regression Model with Segmented Relationship(s)*** Call: segmented.lm(obj = linearFit) Estimated Break-Point(s): Est. St.Err psi1.DayNum 378.498 15.371 Coefficients of the linear terms: Estimate Std. Error t value Pr(>|t|) (Intercept) 122569.14 1286.74 95.25 <2e-16 *** DayNum 698.58 19.20 36.38 <2e-16 *** U1.DayNum 481.42 20.39 23.62 NA --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 Residual standard error: 5559 on 80 degrees of freedom Multiple R-Squared: 0.9996, Adjusted R-squared: 0.9996 Boot restarting based on 9 samples. Last fit: Convergence attained in 2 iterations (rel. change 4.3618e-11) Est. CI(95%).low CI(95%).up psi1.DayNum 378.498 347.908 409.089 $DayNum Est. St.Err. t value CI(95%).l CI(95%).u slope1 698.58 19.2040 36.377 660.37 736.8 slope2 1180.00 6.8399 172.520 1166.40 1193.6 - Segmented fit, testFold = 2 o Train data: 84 points o Test data: 42 points o Simple linear fit, no kinks: Call: lm(formula = Soldiers ~ DayNum, data = trainData) Residuals: Min 1Q Median 3Q Max -100436 -7857 4087 10945 61867 Coefficients: Estimate Std. Error t value Pr(>|t|) (Intercept) 1.033e+05 2.509e+03 41.15 <2e-16 *** DayNum 1.007e+03 8.733e+00 115.35 <2e-16 *** --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 Residual standard error: 20330 on 82 degrees of freedom Multiple R-squared: 0.9939, Adjusted R-squared: 0.9938 F-statistic: 1.331e+04 on 1 and 82 DF, p-value: < 2.2e-16 o Davies test for need of kink: Davies' test for a change in the slope data: formula = Soldiers ~ DayNum , method = lm model = gaussian , link = identity segmented variable = DayNum 'best' at = 390.67, n.points = 10, p-value < 2.2e-16 alternative hypothesis: two.sided o Segmented fit: ***Regression Model with Segmented Relationship(s)*** Call: segmented.lm(obj = linearFit) Estimated Break-Point(s): Est. St.Err psi1.DayNum 386.29 16.533 Coefficients of the linear terms: Estimate Std. Error t value Pr(>|t|) (Intercept) 122525.07 1392.84 87.97 <2e-16 *** DayNum 699.43 20.92 33.43 <2e-16 *** U1.DayNum 487.47 22.22 21.94 NA --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 Residual standard error: 6058 on 80 degrees of freedom Multiple R-Squared: 0.9995, Adjusted R-squared: 0.9994 Boot restarting based on 9 samples. Last fit: Convergence attained in 2 iterations (rel. change 3.0715e-12) Est. CI(95%).low CI(95%).up psi1.DayNum 386.29 353.389 419.191 $DayNum Est. St.Err. t value CI(95%).l CI(95%).u slope1 699.43 20.9210 33.432 657.8 741.06 slope2 1186.90 7.4731 158.820 1172.0 1201.80 - Segmented fit, testFold = 3 o Train data: 84 points o Test data: 42 points o Simple linear fit, no kinks: Call: lm(formula = Soldiers ~ DayNum, data = trainData) Residuals: Min 1Q Median 3Q Max -96274 -7992 4130 10648 48586 Coefficients: Estimate Std. Error t value Pr(>|t|) (Intercept) 1.040e+05 2.504e+03 41.53 <2e-16 *** DayNum 9.963e+02 9.112e+00 109.34 <2e-16 *** --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 Residual standard error: 20220 on 82 degrees of freedom Multiple R-squared: 0.9932, Adjusted R-squared: 0.9931 F-statistic: 1.196e+04 on 1 and 82 DF, p-value: < 2.2e-16 o Davies test for need of kink: Davies' test for a change in the slope data: formula = Soldiers ~ DayNum , method = lm model = gaussian , link = identity segmented variable = DayNum 'best' at = 390, n.points = 10, p-value < 2.2e-16 alternative hypothesis: two.sided o Segmented fit: ***Regression Model with Segmented Relationship(s)*** Call: segmented.lm(obj = linearFit) Estimated Break-Point(s): Est. St.Err psi1.DayNum 386.076 15.959 Coefficients of the linear terms: Estimate Std. Error t value Pr(>|t|) (Intercept) 122457.02 1333.72 91.82 <2e-16 *** DayNum 700.73 20.17 34.74 <2e-16 *** U1.DayNum 481.46 21.48 22.41 NA --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 Residual standard error: 5836 on 80 degrees of freedom Multiple R-Squared: 0.9994, Adjusted R-squared: 0.9994 Boot restarting based on 6 samples. Last fit: Convergence attained in 2 iterations (rel. change 2.2969e-10) Est. CI(95%).low CI(95%).up psi1.DayNum 386.076 354.316 417.836 $DayNum Est. St.Err. t value CI(95%).l CI(95%).u slope1 700.73 20.1720 34.738 660.58 740.87 slope2 1182.20 7.3944 159.880 1167.50 1196.90 - Segmented fit, testFold = NA o Train data: 126 points o Test data: 126 points o Simple linear fit, no kinks: Call: lm(formula = Soldiers ~ DayNum, data = trainData) Residuals: Min 1Q Median 3Q Max -99961 -8039 4059 10899 63488 Coefficients: Estimate Std. Error t value Pr(>|t|) (Intercept) 1.034e+05 2.012e+03 51.38 <2e-16 *** DayNum 1.006e+03 7.012e+00 143.47 <2e-16 *** --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 Residual standard error: 19960 on 124 degrees of freedom Multiple R-squared: 0.994, Adjusted R-squared: 0.994 F-statistic: 2.058e+04 on 1 and 124 DF, p-value: < 2.2e-16 o Davies test for need of kink: Davies' test for a change in the slope data: formula = Soldiers ~ DayNum , method = lm model = gaussian , link = identity segmented variable = DayNum 'best' at = 419, n.points = 10, p-value < 2.2e-16 alternative hypothesis: two.sided o Segmented fit: ***Regression Model with Segmented Relationship(s)*** Call: segmented.lm(obj = linearFit) Estimated Break-Point(s): Est. St.Err psi1.DayNum 383.971 12.935 Coefficients of the linear terms: Estimate Std. Error t value Pr(>|t|) (Intercept) 122516.84 1086.13 112.80 <2e-16 *** DayNum 699.58 16.32 42.88 <2e-16 *** U1.DayNum 483.74 17.33 27.92 NA --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 Residual standard error: 5785 on 122 degrees of freedom Multiple R-Squared: 0.9995, Adjusted R-squared: 0.9995 Boot restarting based on 6 samples. Last fit: Convergence attained in 2 iterations (rel. change 2.3557e-10) Est. CI(95%).low CI(95%).up psi1.DayNum 383.971 358.366 409.576 $DayNum Est. St.Err. t value CI(95%).l CI(95%).u slope1 699.58 16.3170 42.875 667.28 731.88 slope2 1183.30 5.8359 202.760 1171.80 1194.90 * Crossvalidation and final whole-dataset fit results: TestFold Kink sd.Kink. Slope1 sd.Slope1. Slope2 sd.Slope2. lm.Adj.R2 1 1 378.498 15.371 698.58 19.204 1180.0 6.840 0.995 2 2 386.290 16.533 699.43 20.921 1186.9 7.473 0.994 3 3 386.076 15.959 700.73 20.172 1182.2 7.394 0.993 4 NA 383.971 12.935 699.58 16.317 1183.3 5.836 0.994 Adj.R2 lm.RMSE RMSE 1 1.000 21409.94 6225.298 2 0.999 19225.40 5260.739 3 0.999 19936.96 5722.795 4 0.999 19799.92 5692.288 * Segmented Analysis of Russian Casualties in Ukraine completed Fri Sep 11 16:44:29 2026 (0.4 sec elapsed).