1 / 34

Demographic Forecasting and the Role of Priors

This lecture explores the machine learning problem of demographic forecasting, specifically focusing on the Bayesian regularization framework and the use of priors. It discusses the importance of accurate forecasting in various applications such as pension planning, allocation of public health resources, manpower needs planning, and guidance for epidemiological studies. The lecture also delves into the challenges of forecasting very short time series and highlights the use of lagged covariates in regression models. Additionally, it explores the concept of priors and how they can be constructed based on prior knowledge of mortality age profiles and the smoothness of the data. Examples and demos are provided to illustrate the application of these concepts.

ruby
Download Presentation

Demographic Forecasting and the Role of Priors

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Demographic Forecasting andthe Role of Priors Federico Girosi The RAND Corporation Santa Monica, CA, USA

  2. Reference All the material for this lecture can be found at http://gking.harvard.edu/files/smooth/

  3. Plan of the Lecture Demographic forecasting is a machine learning problem Solving the problem in the Bayesian/ regularization framework A closer look at one dimensional priors A closer look at the smoothness parameter Examples/Demos

  4. Forecasting Mortality and Disease Burden Has Important Applications Pension planning Allocation of public health resources Planning manpower needs Guidance for epidemiological studies

  5. Problem: forecasting very short time series

  6. : Mortality in country c, age a and time t : Regression coefficient : Lagged covariates The forecasting problem is set as a regression problem

  7. : Lagged covariates Typical Lagged Covariates GDP Human capital Fat consumption Water quality Cigarette consumption

  8. In most cases some “pooling” is necessary Regressions cannot be estimated separately across age groups or countries. 17 separate regressions (one for each age group) Bad idea!

  9. Those who have knowledge do not predict. Those who predict do not have knowledge Lao Tzu, 6th century BC

  10. The Standard Bayesian Approach Likelihood: Prior Posterior

  11. A Way Out We do need some sort of prior on the β ... but we do not really have prior knowledge on β ... BUT we do have knowledge on μ! AND μ is related to β: μ = X β

  12. Strategy to build a prior Define a non-parametric prior for μ, as a function of the cross-sectional index (age, for example) Use the relationship between μ and β (μ = X β) to change variables and obtain a prior for β

  13. What type of prior knowledge? Mortality age profiles are smooth deformations of well known shapes Mortality varies smoothly across countries Mortality varies smoothly over time

  14. A Good Prior on μ Discretizing age on a grid:

  15. Only a Step Away from Prior on β … The matrix W is fully determined by the order of the derivative n The “template” age profile μ can be made disappear by subtracting if from the data Just need to substitute the specification μ = X β _

  16. And the Prior for β is: where

  17. But What Does the Prior Really Mean?

  18. But What Does the Prior Really Mean? Discretizing over age and fixing one year in time μ is simply a vector of random variables How do the samples from this prior look like?

  19. Demos Samples from prior with zero mean Samples from prior with non zero mean

  20. And what is the role of λ? Two important, related identities

  21. The role of λ λ determines the size of the smoothness functional λ determines the average standard deviation of the prior

  22. Demos Samples from prior with non zero mean: varying the smoothness parameter

  23. Other Types of Priors Time Time trends over age

  24. Dealing with Multiple Smoothness Parameters • Writing the priors is easy … • Estimating the 3 smoothing parameters is very difficult • Cross validation is hard to do with very short time series • Some prior knowledge on the smoothing parameters is needed

  25. Estimating the smoothness parameters Key observation: the smoothness parameters control ALL expected values of the prior

  26. Estimating the smoothness parameters • Sometimes we do have other forms of prior knowledge • How much the dependent variables changes from one cross section (or year) to the next

  27. Estimating the smoothness parameters Expected values of any function of μ can be estimated empirically, by sampling the prior The following equations can be solved numerically:

  28. Demo: Deaths by Transportation Accidents in Chile

  29. Transportation Accidents: no pooling

  30. Pooling Over Countries:Transportation Accidents in Argentina No Pooling Pooling

  31. Summary Regularization theory is a powerful framework that reaches beyond standard pattern recognition In some application it is important to pay attention to the precise nature of the prior Prior knowledge applies to the smoothness parameter too

  32. Mortality age profiles are well known andconsistent across countries and time Back

  33. “Similar” countries have similar mortality patterns Greece France Cyprus Italy Israel Spain Chile Back

  34. Before and After the CureRespiratory Infections in Belize Back

More Related