Diagnostic analysis and computational strategies for estimating discrete time duration models—A Monte Carlo study

Li, Xianghong; Smith, Barry

doi:10.1016/j.jeconom.2015.02.024

Cited by 8 publications

(3 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This algorithm searches a wide range of the likelihood surface before it narrows down the area over which it moves to the nearest maximum. SeeGoffe (1996) andLi and Smith (2010) for an application for duration models.…”

mentioning

confidence: 99%

Ethnic concentration and language fluency of immigrants: Evidence from the guest-worker placement in Germany

Danzer

Yaman

2016

Journal of Economic Behavior & Organization

View full text Add to dashboard Cite

This paper analyses the impact of regional ethnic concentration on the language proficiency and language use of immigrants. It solves the endogeneity of immigrants' location choices by exploiting a peculiar episode of the German immigration history: the exogenous placement of guest-workers after WWII, one of the largest guest-worker programs on record. The econometric approach accounts for several sources of measurement error and provides a falsification exercise that hypothetically relocates the most language proficient immigrants into ethnic enclaves to test the extent of cross-regional sorting necessary to render the results purely spurious. The results show a robust negative effect of ethnic concentration on immigrants' language ability which is driven by differences in contact rates with natives and not by differences in the willingness to integrate. Simple policy simulations suggest education-based entry restrictions as powerful tools for fostering language acquisition and the integration of immigrants.

show abstract

mentioning

confidence: 99%

Ethnic concentration and language fluency of immigrants: Evidence from the guest-worker placement in Germany

Danzer

Yaman

2016

Journal of Economic Behavior & Organization

View full text Add to dashboard Cite

show abstract

“…[1] is tight and effective and the FORTRAN implementation in [2] is dependable, easily extensible and sufficiently fast that it can be applied to both complicated small-scale problems and to very large Monte Carlo studies such as the one in Ref. [3].…”

Section: Sa and Penalty-constrained Optimizationmentioning

confidence: 99%

Simulated Annealing of Constrained Statistical Functions

Smith¹

2017

Computational Optimization in Engineering - Paradigms and Applications

Self Cite

View full text Add to dashboard Cite

In 1987, Corana et al. published a simulated annealing (SA) algorithm. Soon thereafter in 1993, Goffe et al. coded the algorithm in FORTRAN and showed that SA could uncover global optima missed by traditional optimization software when applied to statistical modeling and estimation in economics (econometrics). This chapter shows how and why SA can be used successfully to perform likelihood-based statistical inference on models where likelihood is constrained by often very complicated functions defined on a compact parameter space. These constraints arise because likelihood-based inference involves comparing the maxima of constrained versus unconstrained statistical optimization models. The chapter begins with a review of the relevant literature on SA and constrained optimization using penalty techniques. Next, a constrained optimization problem based in maximum likelihood stress-strength modeling is introduced, and its statistical and numerical properties are summarized. SA is then used to solve a sequence of penalty-constrained optimization problems, and the results are used to construct a confidence interval for the parameter of interest in the statistical model. The chapter concludes with a brief summary of the results and some ways we were able to enhance the performance of SA in this setting.

show abstract

“…Estimation thus needs to be carried out by global optimization routines, which can take long times to conclude without a guarantee of having found the global maximum. We have estimated the FIML model by simulated annealing (see Goffe, Ferrier and Rogers , and Li and Smith ). In order to reduce the computing time, we do not use the time varying nature of our covariates.…”

Section: Estimationmentioning

confidence: 99%

Does Homeownership Prolong the Duration of Unemployment?

Taskin¹,

Yaman

2016

Real Estate Economics

View full text Add to dashboard Cite

We examine the effects of homeownership on individuals' unemployment durations. An unemployment spell can terminate with a job or with nonparticipation. The endogeneity of homeownership is addressed by estimating a full maximum likelihood function jointly modeling the competing hazards and the probability of being a homeowner. Unobserved factors contributing to the probability of being a homeowner are allowed to be correlated with unobservable heterogeneity in the hazard rates. Not controlling for ownership selection, there is neither a significant difference in the job‐finding hazard nor in the nonparticipation hazard of unemployed owners and renters. If we jointly model the ownership selection, we find that unemployed homeowners are more likely to find a job than renters.

show abstract

Diagnostic analysis and computational strategies for estimating discrete time duration models—A Monte Carlo study

Cited by 8 publications

References 36 publications

Ethnic concentration and language fluency of immigrants: Evidence from the guest-worker placement in Germany

Ethnic concentration and language fluency of immigrants: Evidence from the guest-worker placement in Germany

Simulated Annealing of Constrained Statistical Functions

Does Homeownership Prolong the Duration of Unemployment?

Contact Info

Product

Resources

About