Predicting Probabilities and Outcomes with Estimated Models

Once a model has been estimated, it can be used to predict probabilities and / or outcomes for a set of alternatives. This vignette demonstrates examples of how to so using the predict() method along with an estimated model.

You can make predictions for any set of alternatives, so long as the columns in the alternatives correspond to estimated coefficients in your model. By default, if no new data are provided via the newdata argument, then predictions will be made for the original data used to estimate the model.

Predictions can be made using both preference space and WTP space models, as well as multinomial logit and mixed logit models. For mixed logit models, heterogeneity is modeled by simulating draws from the population estimates of the estimated model.

Predicting probabilities

Preference space models

In the example below, a preference space MNL model is estimated (mnl_pref) and then used to predict probabilities for the data used to estimate the model:

library("logitr")

mnl_pref <- logitr(
  data    = yogurt,
  outcome = 'choice',
  obsID   = 'obsID',
  pars    = c('price', 'feat', 'brand')
)

probs <- predict(mnl_pref)
head(probs)
#>   obsID predicted_prob
#> 1     1     0.41802407
#> 2     1     0.02118240
#> 3     1     0.23691737
#> 4     1     0.32387615
#> 5     2     0.26643822
#> 6     2     0.02255486

The predict() method returns a data frame containing the observation ID as well as the predicted probabilities. The original data can also be returned in the data frame by setting returnData = TRUE:

probs <- predict(mnl_pref, returnData = TRUE)
head(probs)
#>   obsID predicted_prob price feat brandhiland brandweight brandyoplait choice
#> 1     1     0.41802407   8.1    0           0           0            0      0
#> 2     1     0.02118240   6.1    0           1           0            0      0
#> 3     1     0.23691737   7.9    0           0           1            0      1
#> 4     1     0.32387615  10.8    0           0           0            1      0
#> 5     2     0.26643822   9.8    0           0           0            0      1
#> 6     2     0.02255486   6.4    0           1           0            0      0

To make predictions for a new set of alternatives, use the newdata argument. The example below makes predictions for just two of the choice observations from the yogurt dataset:

data <- subset(
  yogurt, obsID %in% c(42, 13),
  select = c('obsID', 'alt', 'price', 'feat', 'brand'))

probs_mnl_pref <- predict(
  mnl_pref,
  newdata = data,
  obsID = "obsID",
)

probs_mnl_pref
#>   obsID predicted_prob
#> 1    13     0.43685145
#> 2    13     0.03312986
#> 3    13     0.19155548
#> 4    13     0.33846321
#> 5    42     0.60764778
#> 6    42     0.02602007
#> 7    42     0.17803313
#> 8    42     0.18829902

The ci argument can be used to obtain upper and lower bounds of a confidence interval for predicted probabilities, which are estimated using the Krinsky and Robb parametric bootstrapping method (Krinsky and Robb 1986). For example, a 95% CI is obtained with ci = 0.95:

probs_mnl_pref <- predict(
  mnl_pref,
  newdata = data,
  obsID = "obsID",
  ci = 0.95
)

probs_mnl_pref
#>   obsID predicted_prob predicted_prob_lower predicted_prob_upper
#> 1    13     0.43685145           0.41490293           0.45807457
#> 2    13     0.03312986           0.02605820           0.04220765
#> 3    13     0.19155548           0.17660037           0.20716517
#> 4    13     0.33846321           0.31736913           0.35773877
#> 5    42     0.60764778           0.57204770           0.64057358
#> 6    42     0.02602007           0.01872655           0.03732723
#> 7    42     0.17803313           0.16220968           0.19451696
#> 8    42     0.18829902           0.16778560           0.20931234

WTP space models

WTP space models can also be used to predict probabilities. In the example below, a WTP space model is estimated and used to predict probabilities for the same data data frame as in the previous examples:

mnl_wtp <- logitr(
  data     = yogurt,
  outcome  = 'choice',
  obsID    = 'obsID',
  pars     = c('feat', 'brand'),
  scalePar = 'price',
  numMultiStarts = 10
)

probs_mnl_wtp <- predict(
  mnl_wtp,
  newdata = data,
  obsID   = "obsID",
  ci      = 0.95
)

probs_mnl_wtp
#>   obsID predicted_prob predicted_prob_lower predicted_prob_upper
#> 1    13     0.43686141           0.41626125           0.45759058
#> 2    13     0.03312947           0.02631329           0.04204450
#> 3    13     0.19154829           0.17587351           0.20824388
#> 4    13     0.33846083           0.31824968           0.35810533
#> 5    42     0.60767120           0.57426109           0.63738465
#> 6    42     0.02601800           0.01817677           0.03564045
#> 7    42     0.17802363           0.16251472           0.19536331
#> 8    42     0.18828717           0.16746411           0.20889147

Here is a bar chart comparing the predicted probabilities from the preference space and WTP space models. Since both models are equivalent except in different spaces, the predicted probabilities are identical:

library("ggplot2")

probs <- rbind(probs_mnl_pref, probs_mnl_wtp)
probs$model <- c(rep("mnl_pref", 8), rep("mnl_wtp", 8))
probs$alt <- rep(c("dannon", "hiland", "weight", "yoplait"), 4)
probs$obs <- paste0("Observation ID: ", probs$obsID)
ggplot(probs, aes(x = alt, y = predicted_prob, fill = model)) +
    geom_bar(stat = 'identity', width = 0.7, position = "dodge") +
    geom_errorbar(aes(ymin = predicted_prob_lower, ymax = predicted_prob_upper),
                  width = 0.2, position = position_dodge(width = 0.7)) +
    facet_wrap(vars(obs)) +
    scale_y_continuous(limits = c(0, 1)) +
    labs(x = 'Alternative', y = 'Expected Choice Probabilities') +
    theme_bw()

Predicting outcomes

The predict() method can also be used to predict outcomes by setting type = "outcome" (the default is "prob" for predicting probabilities). In the examples below, outcomes are predicted using the same preference space and WTP space models as in the previous examples. The returnData argument is also set to TRUE so that the predicted outcomes can be compared to the actual choices made:

outcomes_pref <- predict(mnl_pref, type = "outcome", returnData = TRUE)
head(outcomes_pref)
#>   obsID predicted_outcome price feat brandhiland brandweight brandyoplait
#> 1     1                 1   8.1    0           0           0            0
#> 2     1                 0   6.1    0           1           0            0
#> 3     1                 0   7.9    0           0           1            0
#> 4     1                 0  10.8    0           0           0            1
#> 5     2                 0   9.8    0           0           0            0
#> 6     2                 0   6.4    0           1           0            0
#>   choice
#> 1      0
#> 2      0
#> 3      1
#> 4      0
#> 5      1
#> 6      0

outcomes_wtp <- predict(mnl_wtp, type = "outcome", returnData = TRUE)
head(outcomes_wtp)
#>   obsID predicted_outcome feat brandhiland brandweight brandyoplait scalePar
#> 1     1                 1    0           0           0            0      8.1
#> 2     1                 0    0           1           0            0      6.1
#> 3     1                 0    0           0           1            0      7.9
#> 4     1                 0    0           0           0            1     10.8
#> 5     2                 1    0           0           0            0      9.8
#> 6     2                 0    0           1           0            0      6.4
#>   choice
#> 1      0
#> 2      0
#> 3      1
#> 4      0
#> 5      1
#> 6      0

The accuracy of each model can be computed by dividing the number of correctly predicted choices by the total number of choices:

chosen_pref <- subset(outcomes_pref, choice == 1)
chosen_pref$correct <- chosen_pref$choice == chosen_pref$predicted_outcome
accuracy_pref <- sum(chosen_pref$correct) / nrow(chosen_pref)
accuracy_pref
#> [1] 0.3888889

chosen_wtp <- subset(outcomes_wtp, choice == 1)
chosen_wtp$correct <- chosen_wtp$choice == chosen_wtp$predicted_outcome
accuracy_wtp <- sum(chosen_wtp$correct) / nrow(chosen_wtp)
accuracy_wtp
#> [1] 0.3909619

These results show that both models correctly predicted choice for approximately 39% of the observations in the yogurt data frame, which is significantly better than random (25%).

References

Krinsky, Itzhak, and A Leslie Robb. 1986. “On Approximating the Statistical Properties of Elasticities.” The Review of Economics and Statistics, 715–19.