Add the beta-delta discount function of Laibson (1997).
V = beta * delta ^ delay
beta is the intercept for low delays (noting that V=1 for delay=0) makes sense to constrain beta>0
delta is the slope and typically is quite close to 1 (delta<1 for discounting)
Priors over beta and delta are taken from Franck et al (2014), Table 2. As always, users are strongly advised to check the priors are appropriate for their experiment/modelling context.
[x] add separate model
[x] add mixed model
[x] add hierarchical model
[x] tidy up help comments in new classes
[x] add to tests
[x] add to wiki
References
Franck, C. T., Koffarnus, M. N., House, L. L., & Bickel, W. K. (2014). Accurate characterization of delay discounting: A multiple model approach using approximate bayesian model selection and a unified discounting measure. Journal of the Experimental Analysis of Behavior, 103(1), 218–233. http://doi.org/10.1002/jeab.128
Add the beta-delta discount function of Laibson (1997).
beta
is the intercept for low delays (noting thatV=1
fordelay=0
) makes sense to constrainbeta>0
delta
is the slope and typically is quite close to 1 (delta<1
for discounting)Priors over
beta
anddelta
are taken from Franck et al (2014), Table 2. As always, users are strongly advised to check the priors are appropriate for their experiment/modelling context.References
Franck, C. T., Koffarnus, M. N., House, L. L., & Bickel, W. K. (2014). Accurate characterization of delay discounting: A multiple model approach using approximate bayesian model selection and a unified discounting measure. Journal of the Experimental Analysis of Behavior, 103(1), 218–233. http://doi.org/10.1002/jeab.128
Laibson, D. (1997). Golden Eggs and Hyperbolic Discounting. The Quarterly Journal of Economics, 112(2), 443–478. http://doi.org/10.1162/003355397555253