Description Usage Arguments Details Value Author(s) References Examples

This function computes the Qini coefficient from a `performance`

object (as created by the function `performance`

).

1 2 |

`x` |
an object of class |

`direction` |
possible values are |

`plotit` |
plot the incremental gains from the fitted model? |

`...` |
additional arguments passed to |

Qini coefficients represent a natural generalizations of the Gini coefficient to the case of uplift. Qini is defined as the area between the actual incremental gains curve from the fitted model and the area under the diagonal corresponding to random targeting. See the references for details.

A list with the following components

`Qini` |
the Qini coefficient as defined above. |

`inc.gains` |
the incremental gain values from the fitted model. |

`random.inc.gains` |
the random incremental gains. |

Leo Guelman <leo.guelman@gmail.com>

Radcliffe, N. and Surry, P. (2011). Real-World Uplift Modelling with Significance-Based Uplift Trees. Portrait Technical Report, TR-2011-1.

Radcliffe, N. (2007). Using control groups to target on predicted lift: Building and assessing uplift models. Direct Marketing Analytics Journal, An Annual Publication from the Direct Marketing Association Analytics Council, pages 14-21.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 | ```
library(uplift)
### simulate data for uplift modeling
set.seed(123)
dd <- sim_pte(n = 1000, p = 20, rho = 0, sigma = sqrt(2), beta.den = 4)
dd$treat <- ifelse(dd$treat == 1, 1, 0)
### fit uplift random forest
fit1 <- upliftRF(y ~ X1 + X2 + X3 + X4 + X5 + X6 + trt(treat),
data = dd,
mtry = 3,
ntree = 100,
split_method = "KL",
minsplit = 200, # need small trees as there is strong uplift effects in the data
verbose = TRUE)
print(fit1)
summary(fit1)
### predict on new data
dd_new <- sim_pte(n = 2000, p = 20, rho = 0, sigma = sqrt(2), beta.den = 4)
dd_new$treat <- ifelse(dd_new$treat == 1, 1, 0)
pred <- predict(fit1, dd_new)
### evaluate model performance
perf <- performance(pred[, 1], pred[, 2], dd_new$y, dd_new$treat, direction = 1)
### compute Qini coefficient
Q <- qini(perf, plotit = TRUE)
Q
``` |

```
Loading required package: RItools
Loading required package: SparseM
Attaching package: 'SparseM'
The following object is masked from 'package:base':
backsolve
Loading required package: MASS
Loading required package: coin
Loading required package: survival
Loading required package: tables
Loading required package: Hmisc
Loading required package: lattice
Loading required package: Formula
Loading required package: ggplot2
Attaching package: 'Hmisc'
The following objects are masked from 'package:base':
format.pval, round.POSIXt, trunc.POSIXt, units
Loading required package: penalized
Welcome to penalized. For extended examples, see vignette("penalized").
uplift: status messages enabled; set "verbose" to false to disable
upliftRF: starting. Wed Dec 13 08:35:06 2017
10 out of 100 trees so far...
20 out of 100 trees so far...
30 out of 100 trees so far...
40 out of 100 trees so far...
50 out of 100 trees so far...
60 out of 100 trees so far...
70 out of 100 trees so far...
80 out of 100 trees so far...
90 out of 100 trees so far...
Call:
upliftRF(formula = y ~ X1 + X2 + X3 + X4 + X5 + X6 + trt(treat),
data = dd, mtry = 3, ntree = 100, split_method = "KL", minsplit = 200,
verbose = TRUE)
Uplift random forest
Number of trees: 100
No. of variables tried at each split: 3
Split method: KL
$call
upliftRF(formula = y ~ X1 + X2 + X3 + X4 + X5 + X6 + trt(treat),
data = dd, mtry = 3, ntree = 100, split_method = "KL", minsplit = 200,
verbose = TRUE)
$importance
var rel.imp
1 X1 39.97286
2 X2 25.07182
3 X4 18.37845
4 X3 16.57687
$ntree
[1] 100
$mtry
[1] 3
$split_method
[1] "KL"
attr(,"class")
[1] "summary.upliftRF"
$Qini
[1] 0.104765
$inc.gains
[1] 0.068 0.117 0.140 0.176 0.185 0.156 0.139 0.117 0.086 0.023
$random.inc.gains
[1] 0.0023 0.0046 0.0069 0.0092 0.0115 0.0138 0.0161 0.0184 0.0207 0.0230
```

Embedding an R snippet on your website

Add the following code to your website.

For more information on customizing the embed code, read Embedding Snippets.