* Segmented Analysis of Russian Casualties in Ukraine on Thu Nov 12 21:11:06 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-2026-11-12.tsv - Found 127 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: 43 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: 85 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 -107128 -8247 4058 11184 80300 Coefficients: Estimate Std. Error t value Pr(>|t|) (Intercept) 1.017e+05 2.759e+03 36.86 <2e-16 *** DayNum 1.026e+03 8.553e+00 119.95 <2e-16 *** --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 Residual standard error: 22570 on 83 degrees of freedom Multiple R-squared: 0.9943, Adjusted R-squared: 0.9942 F-statistic: 1.439e+04 on 1 and 83 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 = 444, 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 394.473 17.14 Coefficients of the linear terms: Estimate Std. Error t value Pr(>|t|) (Intercept) 122525.07 1476.37 82.99 <2e-16 *** DayNum 699.43 22.18 31.54 <2e-16 *** U1.DayNum 500.34 23.19 21.58 NA --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 Residual standard error: 6422 on 81 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 3.9066e-11) Est. CI(95%).low CI(95%).up psi1.DayNum 394.473 360.371 428.576 $DayNum Est. St.Err. t value CI(95%).l CI(95%).u slope1 699.43 22.1760 31.54 655.31 743.55 slope2 1199.80 6.7699 177.22 1186.30 1213.20 - Segmented fit, testFold = 3 o Train data: 85 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 -104484 -8021 5039 11214 89330 Coefficients: Estimate Std. Error t value Pr(>|t|) (Intercept) 1.021e+05 2.824e+03 36.15 <2e-16 *** DayNum 1.019e+03 9.061e+00 112.46 <2e-16 *** --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 Residual standard error: 23060 on 83 degrees of freedom Multiple R-squared: 0.9935, Adjusted R-squared: 0.9934 F-statistic: 1.265e+04 on 1 and 83 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 395.664 16.849 Coefficients of the linear terms: Estimate Std. Error t value Pr(>|t|) (Intercept) 122457.02 1446.00 84.69 <2e-16 *** DayNum 700.73 21.87 32.04 <2e-16 *** U1.DayNum 497.57 22.86 21.77 NA --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 Residual standard error: 6328 on 81 degrees of freedom Multiple R-Squared: 0.9995, Adjusted R-squared: 0.9995 Boot restarting based on 7 samples. Last fit: Convergence attained in 2 iterations (rel. change 4.9226e-11) Est. CI(95%).low CI(95%).up psi1.DayNum 395.664 362.139 429.188 $DayNum Est. St.Err. t value CI(95%).l CI(95%).u slope1 700.73 21.8700 32.04 657.21 744.24 slope2 1198.30 6.6506 180.18 1185.10 1211.50 - Segmented fit, testFold = NA o Train data: 127 points o Test data: 127 points o Simple linear fit, no kinks: Call: lm(formula = Soldiers ~ DayNum, data = trainData) Residuals: Min 1Q Median 3Q Max -104886 -8008 4562 10964 88358 Coefficients: Estimate Std. Error t value Pr(>|t|) (Intercept) 102225.43 2170.68 47.09 <2e-16 *** DayNum 1019.69 6.98 146.09 <2e-16 *** --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 Residual standard error: 21680 on 125 degrees of freedom Multiple R-squared: 0.9942, Adjusted R-squared: 0.9941 F-statistic: 2.134e+04 on 1 and 125 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 = 443.33, 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 390.892 13.476 Coefficients of the linear terms: Estimate Std. Error t value Pr(>|t|) (Intercept) 122516.84 1152.09 106.34 <2e-16 *** DayNum 699.58 17.31 40.42 <2e-16 *** U1.DayNum 494.44 18.17 27.22 NA --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 Residual standard error: 6136 on 123 degrees of freedom Multiple R-Squared: 0.9995, Adjusted R-squared: 0.9995 Boot restarting based on 8 samples. Last fit: Convergence attained in 2 iterations (rel. change 1.0951e-10) Est. CI(95%).low CI(95%).up psi1.DayNum 390.892 364.217 417.567 $DayNum Est. St.Err. t value CI(95%).l CI(95%).u slope1 699.58 17.3070 40.421 665.32 733.84 slope2 1194.00 5.5198 216.310 1183.10 1204.90 * 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 2 394.473 17.140 699.43 22.176 1199.8 6.770 0.994 3 3 395.664 16.849 700.73 21.870 1198.3 6.651 0.993 4 NA 390.892 13.476 699.58 17.307 1194.0 5.520 0.994 Adj.R2 lm.RMSE RMSE 1 1 25812.07 7519.719 2 1 20052.90 5664.114 3 1 18664.52 5796.744 4 1 21511.97 6038.749 * Segmented Analysis of Russian Casualties in Ukraine completed Thu Nov 12 21:11:06 2026 (0.4 sec elapsed).