summary.mqgam.RdTakes a fitted mqgam object produced by qgam::mqgam and produces various useful summaries from it, making use of
the mgcv::summary.gam method.
## S3 method for class 'mqgam'
summary(mqgam)An mqgam object created with qgam::mqgam.
Please see summary.gam for a detailed explanation of the values given.
Please see summary.gam for a detailed explanation of the summaries given.
Fasiolo M., Goude Y., Nedellec R., & Wood S. N. (2017). Fast calibrated additive quantile regression. URL: https://arxiv.org/abs/1707.03307
Wood, S.N. (2011). Fast stable restricted maximum likelihood and marginal likelihood estimation of semiparametric generalized linear models. Journal of the Royal Statistical Society (B) 73(1), 3-36.
# general usage
summary(mqgam = mtqgam_mqgam)
#> $`========================= quantile 0.1 =========================`
#>
#> Family: elf
#> Link function: identity
#>
#> Formula:
#> dependent ~ s(numeric_1) + s(numeric_2) + factor_1 + factor_2 *
#> factor_3
#>
#> Parametric coefficients:
#> Estimate Std. Error z value Pr(>|z|)
#> (Intercept) 11.852 2.550 4.648 3.36e-06 ***
#> factor_1K 0.781 1.274 0.613 0.5400
#> factor_1L 8.450 10.603 0.797 0.4255
#> factor_1M 1.337 1.360 0.983 0.3256
#> factor_2b -5.787 3.029 -1.910 0.0561 .
#> factor_2c -1.783 3.256 -0.548 0.5840
#> factor_2d -2.436 3.023 -0.806 0.4204
#> factor_2e -5.077 2.891 -1.756 0.0791 .
#> factor_3B -2.908 2.419 -1.202 0.2293
#> factor_2b:factor_3B 5.294 3.265 1.621 0.1049
#> factor_2c:factor_3B 1.094 3.474 0.315 0.7529
#> factor_2d:factor_3B 3.149 3.274 0.962 0.3362
#> factor_2e:factor_3B 7.011 3.144 2.230 0.0257 *
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#>
#> Approximate significance of smooth terms:
#> edf Ref.df Chi.sq p-value
#> s(numeric_1) 1.021 1.041 1.662 0.2092
#> s(numeric_2) 3.445 4.269 13.186 0.0115 *
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#>
#> R-sq.(adj) = 0.000297 Deviance explained = 69.2%
#> -REML = 24691 Scale est. = 1 n = 5000
#>
#> $`========================= quantile 0.3 =========================`
#>
#> Family: elf
#> Link function: identity
#>
#> Formula:
#> dependent ~ s(numeric_1) + s(numeric_2) + factor_1 + factor_2 *
#> factor_3
#>
#> Parametric coefficients:
#> Estimate Std. Error z value Pr(>|z|)
#> (Intercept) 30.90829 3.28923 9.397 <2e-16 ***
#> factor_1K 0.09366 1.73020 0.054 0.9568
#> factor_1L 9.16071 7.38443 1.241 0.2148
#> factor_1M 1.23581 1.82688 0.676 0.4987
#> factor_2b -5.89874 4.06093 -1.453 0.1463
#> factor_2c -4.01288 4.19548 -0.956 0.3388
#> factor_2d -3.45826 3.90769 -0.885 0.3762
#> factor_2e -6.78947 4.15173 -1.635 0.1020
#> factor_3B -3.30077 3.12025 -1.058 0.2901
#> factor_2b:factor_3B 5.17050 4.36099 1.186 0.2358
#> factor_2c:factor_3B 3.76976 4.48158 0.841 0.4003
#> factor_2d:factor_3B 4.07557 4.21973 0.966 0.3341
#> factor_2e:factor_3B 8.53886 4.44078 1.923 0.0545 .
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#>
#> Approximate significance of smooth terms:
#> edf Ref.df Chi.sq p-value
#> s(numeric_1) 1.008 1.015 0.814 0.3716
#> s(numeric_2) 2.120 2.644 10.147 0.0164 *
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#>
#> R-sq.(adj) = 0.000398 Deviance explained = 30.6%
#> -REML = 24236 Scale est. = 1 n = 5000
#>
#> $`========================= quantile 0.5 =========================`
#>
#> Family: elf
#> Link function: identity
#>
#> Formula:
#> dependent ~ s(numeric_1) + s(numeric_2) + factor_1 + factor_2 *
#> factor_3
#>
#> Parametric coefficients:
#> Estimate Std. Error z value Pr(>|z|)
#> (Intercept) 51.9919 2.9191 17.811 <2e-16 ***
#> factor_1K -0.1199 1.5381 -0.078 0.938
#> factor_1L 3.5204 6.8065 0.517 0.605
#> factor_1M 1.0064 1.6229 0.620 0.535
#> factor_2b -4.2839 3.6189 -1.184 0.237
#> factor_2c -3.3268 3.7782 -0.881 0.379
#> factor_2d -2.9186 3.4921 -0.836 0.403
#> factor_2e -4.2748 3.6650 -1.166 0.243
#> factor_3B -2.7146 2.7831 -0.975 0.329
#> factor_2b:factor_3B 3.9441 3.8857 1.015 0.310
#> factor_2c:factor_3B 3.1910 4.0314 0.792 0.429
#> factor_2d:factor_3B 3.2132 3.7707 0.852 0.394
#> factor_2e:factor_3B 5.4905 3.9271 1.398 0.162
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#>
#> Approximate significance of smooth terms:
#> edf Ref.df Chi.sq p-value
#> s(numeric_1) 1.008 1.016 0.572 0.45323
#> s(numeric_2) 2.096 2.615 12.105 0.00682 **
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#>
#> R-sq.(adj) = 0.000799 Deviance explained = 0.376%
#> -REML = 23886 Scale est. = 1 n = 5000
#>
#> $`========================= quantile 0.7 =========================`
#>
#> Family: elf
#> Link function: identity
#>
#> Formula:
#> dependent ~ s(numeric_1) + s(numeric_2) + factor_1 + factor_2 *
#> factor_3
#>
#> Parametric coefficients:
#> Estimate Std. Error z value Pr(>|z|)
#> (Intercept) 73.1379 3.2990 22.170 <2e-16 ***
#> factor_1K -0.3210 1.7582 -0.183 0.855
#> factor_1L -0.2357 6.8615 -0.034 0.973
#> factor_1M 0.9733 1.8528 0.525 0.599
#> factor_2b -3.3366 4.1706 -0.800 0.424
#> factor_2c -3.1010 4.3187 -0.718 0.473
#> factor_2d -2.5485 3.9674 -0.642 0.521
#> factor_2e -2.0474 4.1850 -0.489 0.625
#> factor_3B -2.3422 3.1481 -0.744 0.457
#> factor_2b:factor_3B 3.3776 4.4735 0.755 0.450
#> factor_2c:factor_3B 3.1537 4.6090 0.684 0.494
#> factor_2d:factor_3B 2.6359 4.2848 0.615 0.538
#> factor_2e:factor_3B 2.8761 4.4822 0.642 0.521
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#>
#> Approximate significance of smooth terms:
#> edf Ref.df Chi.sq p-value
#> s(numeric_1) 1.022 1.044 0.148 0.71879
#> s(numeric_2) 1.821 2.274 10.319 0.00936 **
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#>
#> R-sq.(adj) = 0.000425 Deviance explained = 30.2%
#> -REML = 24268 Scale est. = 1 n = 5000
#>
#> $`========================= quantile 0.9 =========================`
#>
#> Family: elf
#> Link function: identity
#>
#> Formula:
#> dependent ~ s(numeric_1) + s(numeric_2) + factor_1 + factor_2 *
#> factor_3
#>
#> Parametric coefficients:
#> Estimate Std. Error z value Pr(>|z|)
#> (Intercept) 91.4252 2.3589 38.757 < 2e-16 ***
#> factor_1K -0.7171 1.3579 -0.528 0.59745
#> factor_1L -12.0658 4.3284 -2.788 0.00531 **
#> factor_1M 0.2290 1.4176 0.162 0.87169
#> factor_2b 1.3651 3.0343 0.450 0.65278
#> factor_2c -1.3422 3.0806 -0.436 0.66306
#> factor_2d -1.5505 2.7862 -0.556 0.57788
#> factor_2e -2.1636 2.8654 -0.755 0.45019
#> factor_3B -1.0444 2.2424 -0.466 0.64137
#> factor_2b:factor_3B -1.7951 3.2600 -0.551 0.58187
#> factor_2c:factor_3B 1.0285 3.3011 0.312 0.75539
#> factor_2d:factor_3B 0.1367 3.0412 0.045 0.96416
#> factor_2e:factor_3B 1.4764 3.1070 0.475 0.63465
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#>
#> Approximate significance of smooth terms:
#> edf Ref.df Chi.sq p-value
#> s(numeric_1) 1.015 1.031 4.148 0.04288 *
#> s(numeric_2) 1.056 1.111 9.575 0.00233 **
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#>
#> R-sq.(adj) = -0.00251 Deviance explained = 68.9%
#> -REML = 24687 Scale est. = 1 n = 5000
#>
# printing just one quantile summary
summary(mqgam = mtqgam_mqgam)[[3]]
#>
#> Family: elf
#> Link function: identity
#>
#> Formula:
#> dependent ~ s(numeric_1) + s(numeric_2) + factor_1 + factor_2 *
#> factor_3
#>
#> Parametric coefficients:
#> Estimate Std. Error z value Pr(>|z|)
#> (Intercept) 51.9919 2.9191 17.811 <2e-16 ***
#> factor_1K -0.1199 1.5381 -0.078 0.938
#> factor_1L 3.5204 6.8065 0.517 0.605
#> factor_1M 1.0064 1.6229 0.620 0.535
#> factor_2b -4.2839 3.6189 -1.184 0.237
#> factor_2c -3.3268 3.7782 -0.881 0.379
#> factor_2d -2.9186 3.4921 -0.836 0.403
#> factor_2e -4.2748 3.6650 -1.166 0.243
#> factor_3B -2.7146 2.7831 -0.975 0.329
#> factor_2b:factor_3B 3.9441 3.8857 1.015 0.310
#> factor_2c:factor_3B 3.1910 4.0314 0.792 0.429
#> factor_2d:factor_3B 3.2132 3.7707 0.852 0.394
#> factor_2e:factor_3B 5.4905 3.9271 1.398 0.162
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#>
#> Approximate significance of smooth terms:
#> edf Ref.df Chi.sq p-value
#> s(numeric_1) 1.008 1.016 0.572 0.45323
#> s(numeric_2) 2.096 2.615 12.105 0.00682 **
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#>
#> R-sq.(adj) = 0.000799 Deviance explained = 0.376%
#> -REML = 23886 Scale est. = 1 n = 5000