将marginaleffects::predictions()的结果与主数据框连接起来？

huangapple
go评论97阅读模式
英文:

Join results of marginaleffects::predictions() back to main df?

问题

我运行了一个lm模型，然后在输出上运行了marginaleffects中的predictions()。我想将这个输出与我输入到lm中的主要数据框连接起来，但我看不到在这种情况下应该使用哪个选项。

有人知道我需要在这里做什么吗？predictions()的输出具有一个rowid，但根据索引连接（可能已更改顺序）似乎是一种冒险的方式。

例如，看下面的代码（来自文档）：

  1. mod <- lm(mpg ~ hp + factor(cyl), data = mtcars)
  2. pred <- predictions(mod)
  4. pred %>% head()
  6.  Estimate Std. Error    z Pr(>|z|) 2.5 % 97.5 %
  7.      20.0      1.204 16.6   <0.001  17.7   22.4
  8.      20.0      1.204 16.6   <0.001  17.7   22.4
  9.      26.4      0.962 27.5   <0.001  24.5   28.3
  10.      20.0      1.204 16.6   <0.001  17.7   22.4
  11.      15.9      0.992 16.0   <0.001  14.0   17.9
  12.      20.2      1.219 16.5   <0.001  17.8   22.5
  14. Columns: rowid, estimate, std.error, statistic, p.value, conf.low, conf.high, mpg, hp, cyl 
  16. > mtcars %>% head()
  17.                    mpg cyl disp  hp drat    wt  qsec vs am gear carb
  18. Mazda RX4         21.0   6  160 110 3.90 2.620 16.46  0  1    4    4
  19. Mazda RX4 Wag     21.0   6  160 110 3.90 2.875 17.02  0  1    4    4
  20. Datsun 710        22.8   4  108  93 3.85 2.320 18.61  1  1    4    1
  21. Hornet 4 Drive    21.4   6  258 110 3.08 3.215 19.44  1  0    3    1
  22. Hornet Sportabout 18.7   8  360 175 3.15 3.440 17.02  0  0    3    2
  23. Valiant           18.1   6  225 105 2.76 3.460 20.22  1  0    3    1

在pred中，每行mtcars都有1个预测，但我如何将它们连接起来？

英文:

I've run a lm model and then run predictions() from marginaleffects on the output. I want to join this back to by main df (that I fed into the lm) but I can't see what option is the right one to use in this case.

Does anyone know what I need to do here? The output of predictions() has a rowid but joining on an index (which may have changed order) seems like a risky way forward.

For example, take the following code (from the documentation):

  1. mod &lt;- lm(mpg ~ hp + factor(cyl), data = mtcars)
  2. pred &lt;- predictions(mod)
  4. pred %&gt;% head()
  6.  Estimate Std. Error    z Pr(&gt;|z|) 2.5 % 97.5 %
  7.      20.0      1.204 16.6   &lt;0.001  17.7   22.4
  8.      20.0      1.204 16.6   &lt;0.001  17.7   22.4
  9.      26.4      0.962 27.5   &lt;0.001  24.5   28.3
  10.      20.0      1.204 16.6   &lt;0.001  17.7   22.4
  11.      15.9      0.992 16.0   &lt;0.001  14.0   17.9
  12.      20.2      1.219 16.5   &lt;0.001  17.8   22.5
  14. Columns: rowid, estimate, std.error, statistic, p.value, conf.low, conf.high, mpg, hp, cyl 
  17. &gt; mtcars %&gt;% head()
  18.                    mpg cyl disp  hp drat    wt  qsec vs am gear carb
  19. Mazda RX4         21.0   6  160 110 3.90 2.620 16.46  0  1    4    4
  20. Mazda RX4 Wag     21.0   6  160 110 3.90 2.875 17.02  0  1    4    4
  21. Datsun 710        22.8   4  108  93 3.85 2.320 18.61  1  1    4    1
  22. Hornet 4 Drive    21.4   6  258 110 3.08 3.215 19.44  1  0    3    1
  23. Hornet Sportabout 18.7   8  360 175 3.15 3.440 17.02  0  0    3    2
  24. Valiant           18.1   6  225 105 2.76 3.460 20.22  1  0    3    1

There is 1 prediction in pred for each row in mtcars but how do I join these?

答案1

得分: 2

Ah, I got it! It seems you just need to pass the original df through to the newdata argument of predictions. E.g.

  1. mod &lt;- lm(mpg ~ hp + factor(cyl), data = mtcars)
  2. pred &lt;- predictions(mod, newdata = mtcars)
  4. pred %&gt;% head()
  6.  估计值 标准误差    z Pr(>|z|) 2.5 % 97.5 % cyl disp  hp drat   wt qsec vs am gear
  7.      20.0      1.204 16.6   &lt;0.001  17.7   22.4   6  160 110 3.90 2.62 16.5  0  1    4
  8.      20.0      1.204 16.6   &lt;0.001  17.7   22.4   6  160 110 3.90 2.88 17.0  0  1    4
  9.      26.4      0.962 27.5   &lt;0.001  24.5   28.3   4  108  93 3.85 2.32 18.6  1  1    4
  10.      20.0      1.204 16.6   &lt;0.001  17.7   22.4   6  258 110 3.08 3.21 19.4  1  0    3
  11.      15.9      0.992 16.0   &lt;0.001  14.0   17.9   8  360 175 3.15 3.44 17.0  0  0    3
  12.      20.2      1.219 16.5   &lt;0.001  17.8   22.5   6  225 105 2.76 3.46 20.2  1  0    3
  13.  carb
  14.     4
  15.     4
  16.     1
  17.     1
  18.     2
  19.     1
  21. Columns: rowid, estimate, std.error, statistic, p.value, conf.low, conf.high, mpg, cyl, disp, hp, drat, wt, qsec, vs, am, gear, carb
英文:

Ah, I got it! It seems you just need to pass the original df through to the newdata argument of predictions. E.g.

  1. mod &lt;- lm(mpg ~ hp + factor(cyl), data = mtcars)
  2. pred &lt;- predictions(mod, newdata = mtcars)
  4. pred %&gt;% head()
  6.  Estimate Std. Error    z Pr(&gt;|z|) 2.5 % 97.5 % cyl disp  hp drat   wt qsec vs am gear
  7.      20.0      1.204 16.6   &lt;0.001  17.7   22.4   6  160 110 3.90 2.62 16.5  0  1    4
  8.      20.0      1.204 16.6   &lt;0.001  17.7   22.4   6  160 110 3.90 2.88 17.0  0  1    4
  9.      26.4      0.962 27.5   &lt;0.001  24.5   28.3   4  108  93 3.85 2.32 18.6  1  1    4
  10.      20.0      1.204 16.6   &lt;0.001  17.7   22.4   6  258 110 3.08 3.21 19.4  1  0    3
  11.      15.9      0.992 16.0   &lt;0.001  14.0   17.9   8  360 175 3.15 3.44 17.0  0  0    3
  12.      20.2      1.219 16.5   &lt;0.001  17.8   22.5   6  225 105 2.76 3.46 20.2  1  0    3
  13.  carb
  14.     4
  15.     4
  16.     1
  17.     1
  18.     2
  19.     1
  21. Columns: rowid, estimate, std.error, statistic, p.value, conf.low, conf.high, mpg, cyl, disp, hp, drat, wt, qsec, vs, am, gear, carb

huangapple
  • 本文由 发表于 2023年6月18日 21:33:30
  • 转载请务必保留本文链接：https://go.coder-hub.com/76500798.html
  • r
  • r-marginaleffects
匿名

发表评论

匿名网友

:?: :razz: :sad: :evil: :!: :smile: :oops: :grin: :eek: :shock: :???: :cool: :lol: :mad: :twisted: :roll: :wink: :idea: :arrow: :neutral: :cry: :mrgreen:

确定