- freely available
- re-usable
Games 2011, 2(1), 16-20; doi:10.3390/g2010016
Published: 7 February 2011
Abstract
: This note briefly summarizes the consequences of adding correlated individual differences to the best baseline model in the Games competition, I-SAW. I find evidence that the traits of an individual are correlated, but refining I-SAW to capture these correlations does not significantly improve the model's accuracy when predicting average behavior.1. Introduction
Are individual differences correlated and can modeling them as such increase the accuracy of a model's predictions? Correlations between individual differences was one of the features of the model I entered in the Games choice prediction competition (CPC). This note briefly summarizes the consequences of adding correlated individual differences to the I-SAW model, the best baseline model in the CPC. It is assumed the reader is familiar with the experiments, I-SAW model, and competition described by Erev et al. [1].
One of the regularities observed in the results of the CPC estimation experiment and previous studies is that individuals differ in their behavior [2,3]. In the I-SAW model, individual differences in behavior are determined by five trait parameters, which are summarized in Table 1.
In the standard version of I-SAW, it is assumed that the values an individual's trait parameters take are independent. This study is motivated by the conjecture that traits are correlated. For example, one might expect that exploration (ε_{i}), the tendency to choose at random, is negatively correlated with inertia (π_{i}), the tendency to repeat the previous choice. Conversely, one might expect that the tendency of participants to give more weight to average payoffs (w_{i}) is positively correlated with inertia (π_{i}).
Data from the CPC experiments were used to investigate correlations between trait parameters. First, the individual decisions made by the 120 participants in the estimation experiment were used to calculate maximum likelihood estimates of the five trait parameters for each participant. These estimates provide evidence that traits are correlated and not independent. Second, the estimation experiment was simulated with a refined I-SAW model. Parameters were constrained to have the same uniform distributions as in the baseline I-SAW model. Selected correlations between trait parameters were introduced and estimated using a grid search. Interestingly, adding these correlations between trait parameters while holding the model and distribution of parameters constant had only a very small effect on how accurately the model predicts average behavior in the estimation and competition experiments.
2. Estimating Trait Parameters and Identifying Correlations
Entry decisions are stochastic in the I-SAW model. In the first trial, each player enters with a fixed probability. In subsequent trials, the probability player i enters depends on the realized and forgone payoffs in the previous t – 1 trials and on player i's traits. Traits vary between individuals, but for a given individual are constant across trials and decision problems.
Each of the 120 participants in the CPC estimation experiment played 10 games, g, and each game had 50 trials, t. For each participant, the 500 observed entry decisions were used to calculate maximum likelihood estimates of a vector containing the five trait parameters θ_{i} = (ε_{i},π_{i},μ_{i}, ρ_{i}, w_{i}). The following log likelihood function was used (a similar approach has been used by Yechiam and Busemeyer [3,4]):
For exploration and inertia, the exact probability of entry was calculated directly. For exploitation, to accommodate the internal stochastic component of I-SAW (drawing a small sample of previous outcomes of the current game), it was calculated as follows
The vector of parameters θ_{i} was estimated using a nonlinear optimization method. The parameter values were constrained by the upper and lower bounds of the distributions shown in Table 1. The correlation coefficients and summary statistics of the estimates are shown in Table 2. The coefficients shown in the top half of the table suggest an individual's trait parameters are not independent. The following relationships between individual traits are apparent. Participants with a higher tendency to explore ε_{i} have a lower tendency to exhibit inertia π_{i}. They also give less weight to average payoffs over all previous trials w_{i} and more weight to a small sample of previous trials (1 – w_{i}). This would cause them to have a greater tendency to underweight rare events. Participants with a higher tendency for inertia π_{i}, in contrast, give more weight to average payoffs w_{i}. Finally, in sampling during exploitation trials, participants with a greater tendency to bias their sample by selecting the most recent trial ρ_{i}, give less weight to average payoffs w_{i}.
4. Discussion
This study found that while correlations between traits matter for individual behavior, refining I-SAW to capture these correlations does not significantly improve the prediction of averages such as average entry rates, efficiency, and alternation rates. Hence, when the goal is prediction of average rather than individual behavior, assuming individual trait parameters are independently distributed appears to be a sound simplifying assumption. A natural question for future research is why the refinement of I-SAW did not produce better predictions. One possibility is that since participants interacted in groups, there may be group effects that carry over to the parameters obtained by fitting the model separately to each individual. Another direction for future research is testing models with fewer trait parameters. If, as this study suggests, correlations between trait parameters only have a small effect on the accuracy of the model's prediction of average behavior, it may be possible to achieve the same degree of predictive accuracy with a model that is simplified by combining some of the trait parameters.
Table 1. The five trait parameters. |
Parameter and Distribution | Description |
---|---|
ε_{i} ∼ U[0,0.24] | Probability of exploration in trials after the first one. |
π_{i} ∼ U[0,0.6] | Tendency for inertia. |
μ_{i} = {1, 2, or 3} | Number of samples taken in exploitation trials. |
ρ_{i} ∼ U[0,0.2] | Probability a sample draw is biased. If the draw is biased, the most recent trial is selected. If it is unbiased, a previous trial is selected at random. |
w_{i} ∼ U[0,0.8] | In exploitation trials, the sample mean is given weight (1 – w) and the mean of all previous trials weight w. |
Table 2. Correlation coefficients and summary statistics for estimated trait parameters. |
Parameter | ε_{i} | π_{i} | μ_{i} | ρ_{i} | w_{i} |
---|---|---|---|---|---|
ε_{i} | 1.00 | ||||
π_{i} | –0.38*** | 1.00 | |||
μ_{i} | –0.03 | 0.14 | 1.00 | ||
ρ_{i} | –0.22* | –0.14 | –0.14 | 1.00 | |
w_{i} | –0.31*** | 0.24** | 0.05 | –0.27** | 1.00 |
mean | 0.12 | 0.31 | 1.87 | 0.12 | 0.29 |
variance | 0.01 | 0.04 | 0.70 | 0.01 | 0.05 |
max | 0.24 | 0.60 | 3.00 | 0.20 | 0.80 |
min | 0.00 | 0.00 | 1.00 | 0.00 | 0.00 |
^{*}:p < .05,^{**}:p < 0.01,^{***}:p < 0.001
References and Notes
- Erev, I.; Ert, E.; Roth, A. A choice prediction competition for market entry games: An introduction. Games 2010, 1, 117–136.
- Ert, E.; Yechiam, E. Consistent constructs in individuals' risk taking in decisions from experience. Acta Psychol. 2010, 134, 225–232.
- Yechiam, E.; Busemeyer, J.R. Evaluating generalizability and parameter consistency in learning models. Games Econ. Behav. 2008, 63, 370–394.
- Yechiam, E.; Busemeyer, J.R. Comparison of basic assumptions embedded in learning models for experience based decision-making. Psychonomic Bull. 2005, 12, 387–402.
- ^{1}The sum of independent normally distributed random variables has the following property: if ${X}_{1}\sim N({\mathit{\mu}}_{1},{\mathit{\sigma}}_{1}^{2})$ and ${X}_{2}\sim N({\mathit{\mu}}_{2},{\mathit{\sigma}}_{2}^{2})$ then $({X}_{1}+{X}_{2})\sim N({\mathit{\mu}}_{1}+{\mathit{\mu}}_{2},{\mathit{\sigma}}_{1}^{2}+{\mathit{\sigma}}_{2}^{2})$. Further, if X~N(μ, σ^{2}) and Y = aX, then Y~N(aμ, a^{2}σ^{2}). Let x = (x_{1},…x_{n}) be a vector of independent standard normal values. Let Z = ∑_{i}a_{i}x_{i}. If ${\sum}_{i}{a}_{i}^{2}=1$, it follows Z∼N (0,1)
© 2011 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/3.0/).