Testing for IIA in the Multinomial Logit Model

By Simon Cheng, Scott Long. 2007

374 citations recorded by the source corpus.

Subfields: Economics and Econometrics, General Decision Sciences, Marketing

View Paper: https://www.semanticscholar.org/paper/49c395f0c94927f961021be01d8de32086a43d40

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Economics and Econometrics

1. J. Hessels
2. Roy Thurik
3. Peter van der Zwan
4. I. Verheul
5. P. C. Motel
6. J. Choumert
7. Andrew S. Belasco
8. Prashanth Mahagaonkar

In General Decision Sciences

1. J. Hessels
2. P. C. Motel
3. J. Choumert
4. Andrew S. Belasco
5. Prashanth Mahagaonkar
6. Andrés Barge-Gil
7. Devrim Göktepe-Hultén
8. Peter van der Zwan

In Marketing

1. J. Hessels
2. Andrew S. Belasco
3. Prashanth Mahagaonkar
4. Andrés Barge-Gil
5. Devrim Göktepe-Hultén
6. Peter van der Zwan
7. Harald Strotmann
8. Andreas Koch

Citations in the corpus

• Interpreting Results From the Multinomial Logit Model (2015, neutral): “These tests have shown poor properties in small samples and often show conflicting results under certain data structures (Cheng & Long, 2007), but they are still the most common tests available to test the assumption of IIA and recommended in parts of the literature to help choose between the MLM…”
• Stacking up the ladder: A panel data analysis of Tanzanian household energy choices (2017, background): “There has, however, been extensive engagement on the reliability of these types of tests which cast doubt on their usefulness for empirical work (Cheng and Long, 2007; Fry and Harris, 1998; Fry and Harris, 1996).”
• ABSORPTIVE CAPACITY AND INNOVATION IN THE KNOWLEDGE INTENSIVE BUSINESS SERVICE SECTOR (2008, background): “16 This scepticism is reduced, however, if one considers the results from Cheng and Long (2005) based on Monte Carlo simulations.”
• Factors influencing the entrepreneurial engagement of opportunity and necessity entrepreneurs (2016, methodology): “Because of the independence of irrelevant alternatives assumption implicit in multinomial logit models and the problems with testing this assumption in these models (Cheng and Long 2007), we additionally perform a multinomial probit regression.”
• Creating College Opportunity: School Counselors and Their Influence on Postsecondary Enrollment (2013, methodology): “Long (1997) cites two statistical tests commonly used to evaluate IIA: Hausman-McFadden (1984) and Small-Hsaio (1985); however, Long and others, like Cheng and Long (2007), reveal that despite being the best available tests of IIA, both Hasuman-McFadden (HM) and Small-Hsiao (SH) exhibit poor statist”
• Open, Semi-Open and Closed Innovators: Towards an Explanation of Degree of Openness (2010, methodology): “We decided to employ this method since recent studies tend to conclude that IIA tests are not suited to applied work as they often reject the assumption when alternatives seem distinct, and often fail to reject it when alternatives can reasonably be viewed as close substitutes (Chen and Long, 2007 ”
• Inventing and patenting activities of scientists: in the expectation of money or reputation? (2010, methodology): “This dissimilarity is mainly attributed to the method deficiencies in Small– Hsiao tests mentioned by Long and Freese (2006) who cite a Monte-Carlo study by Cheng and Long (2007) concluding that Small–Hsiao tests are not sufficient to test for IIA Inventing and patenting activities of scientists 413”