Discrete Choice Methods with Simulation by Kenneth E. Train

By Kenneth E. Train

Targeting the numerous advances which are made attainable through simulation, this publication describes the hot iteration of discrete selection tools. Researchers use those statistical how you can research the alternatives that buyers, families, organisations, and different brokers make. all of the significant types is roofed: logit, generalized severe price, or GEV (including nested and cross-nested logits), probit, and combined logit, plus various standards that construct on those fundamentals. The strategies are appropriate in lots of fields, together with strength, transportation, environmental experiences, well-being, hard work, and advertising and marketing.

Show description

Read or Download Discrete Choice Methods with Simulation PDF

Best econometrics books

Stochastic Limit Theory: An Introduction for Econometricicans (Advanced Texts in Econometrics)

This significant new econometrics textual content surveys fresh advancements within the swiftly increasing box of asymptotic distribution conception, with a unique emphasis at the difficulties of time dependence and heterogeneity. Designed for econometricians and complex scholars with restricted mathematical education, the e-book sincerely lays out the mandatory math and chance thought and makes use of a number of examples to make its info beneficial and understandable.

Forecasting Non-Stationary Economic Time Series

Economies evolve and are topic to unexpected shifts prompted by means of legislative alterations, financial coverage, significant discoveries, and political turmoil. Macroeconometric types are a really imperfect device for forecasting this hugely advanced and altering strategy. Ignoring those elements results in a large discrepancy among idea and perform.

Economics of Insurance

The idea of coverage is gifted during this booklet, mentioned from the perspective of the idea of economics of uncertainty. the primary of top class calculation which the publication makes use of relies on fiscal equilibrium idea and differs from the various top class platforms mentioned by means of actuaries. Reinsurance is built within the framework of normal financial equilibrium concept lower than uncertainty.


This can be an excerpt from the 4-volume dictionary of economics, a reference ebook which goals to outline the topic of economics this present day. 1300 topic entries within the whole paintings conceal the large topics of monetary conception. This extract concentrates on econometrics.

Additional resources for Discrete Choice Methods with Simulation

Sample text

P1: GEM/IKJ P2: GEM/IKJ QC: GEM/ABE CB495-03Drv CB495/Train KEY BOARDED 48 T1: GEM August 20, 2002 12:14 Char Count= 0 Behavioral Models The limitation of the logit model arises when we attempt to allow tastes to vary with respect to unobserved variables or purely randomly.

This probability is a J-dimensional integral over the density of the J error terms in εn = εn1 , . . , εn J . The dimension can be reduced, however, through recognizing that only differences in utility matter. With J errors (one for each alternative), there are J − 1 error differences. The choice probability can be expressed as a (J − 1)-dimensional integral over the density of these error differences: Pni = Prob(Uni > Un j ∀ j = i) = Prob(εn j − εni < Vni − Vn j ∀ j = i) = Prob(ε˜n ji < Vni − Vn j ∀ j = i) = I (ε˜n ji < Vni − Vn j ∀ j = i)g(ε˜ni ) d ε˜ni P1: GEM/IKJ P2: GEM/IKJ QC: GEM/ABE CB495-02Drv CB495/Train KEY BOARDED T1: GEM September 18, 2002 11:15 Char Count= 0 Properties of Discrete Choice Models 27 where ε˜n ji = εn j − εni is the difference in errors for alternatives i and j; ε˜ni = ε˜n1i , .

The variance of each error difference depends on the variances and covariances of the original errors. For example, the variance of the difference between the first and second errors is Var(ε˜n21 ) = Var(εn2 − εn1 ) = Var(εn1 ) + Var(εn2 ) − 2 Cov(εn1 , εn2 ) = σ11 + σ22 − 2σ12 . We can similarly calculate the covariance between ε˜n21 , which is the difference between the first and second errors, and ε˜n31 , which is the difference between the first and third errors: Cov(ε˜n21 , ε˜n31 ) = E(εn2 − εn1 )(εn3 − εn1 ) = E(εn2 εn3 − εn2 εn1 − εn3 εn1 + εn1 εn1 ) = σ23 − σ21 − σ31 + σ11 .

Download PDF sample

Rated 4.45 of 5 – based on 42 votes