| Titre : |
Techniques of estimation and testing for the limited dependent variable models |
| Type de document : |
texte imprimé |
| Auteurs : |
Hamoudi Hadj Sahraoui, Auteur ; R. W. Blundell, Directeur de thèse ; L. Gill, Directeur de thèse |
| Editeur : |
Manchester : [s.n.] |
| Année de publication : |
1984 |
| Importance : |
72 f. |
| Présentation : |
ill. |
| Format : |
30 cm. |
| Note générale : |
Mémoire de Master : Econometrics and Social Statistics : University of Manchester : 1984
Annexes f. 68-69. - Bibliogr. f. 70-72 |
| Langues : |
Anglais (eng) |
| Mots-clés : |
Econometrics
Logit model
Probit model
Methods likelihood |
| Index. décimale : |
Ms00484 |
| Résumé : |
One of the main objectives of econometrics is the use of statistical and mathematical techniques for the measurement of economic relations and prediction of the economic behaviour of the agents.
Often the phénomenon observed is a continuous one and the econometric theory provides us with a wide range of techniques for estimating this kind of model.
Sometimes, however, one is faced with the problem of the dependent variable being either qualitative or limited in its range.
When the observations on the dependent variable are either "O" or "1" we say that we have a qualitative response data and the model consists of finding the probability that an economic agent with a given set of characteristics will make one choice rather than another.
Two models are used for this kind of qualitative choice model.
The Logit model and the Probit model.
The first section of this dissertation consists mainly of surveying the most used techniques of estimations which I classified in 2 categories: the maximum likelihood procedures and the bias selection methods; I also included the least absolute deviation estimator for censored models for testing purposes.
The remaining section, however will be entirely devoted to surveying the kind of tests which are used to test the validity of the limited dependent variable model. |
Techniques of estimation and testing for the limited dependent variable models [texte imprimé] / Hamoudi Hadj Sahraoui, Auteur ; R. W. Blundell, Directeur de thèse ; L. Gill, Directeur de thèse . - Manchester : [s.n.], 1984 . - 72 f. : ill. ; 30 cm. Mémoire de Master : Econometrics and Social Statistics : University of Manchester : 1984
Annexes f. 68-69. - Bibliogr. f. 70-72 Langues : Anglais ( eng)
| Mots-clés : |
Econometrics
Logit model
Probit model
Methods likelihood |
| Index. décimale : |
Ms00484 |
| Résumé : |
One of the main objectives of econometrics is the use of statistical and mathematical techniques for the measurement of economic relations and prediction of the economic behaviour of the agents.
Often the phénomenon observed is a continuous one and the econometric theory provides us with a wide range of techniques for estimating this kind of model.
Sometimes, however, one is faced with the problem of the dependent variable being either qualitative or limited in its range.
When the observations on the dependent variable are either "O" or "1" we say that we have a qualitative response data and the model consists of finding the probability that an economic agent with a given set of characteristics will make one choice rather than another.
Two models are used for this kind of qualitative choice model.
The Logit model and the Probit model.
The first section of this dissertation consists mainly of surveying the most used techniques of estimations which I classified in 2 categories: the maximum likelihood procedures and the bias selection methods; I also included the least absolute deviation estimator for censored models for testing purposes.
The remaining section, however will be entirely devoted to surveying the kind of tests which are used to test the validity of the limited dependent variable model. |
|
Ms00112 
Ajouter le résultat dans votre panier
Visionner les documents numériques
Faire une suggestion Affiner la rechercheTechniques of estimation and testing for the limited dependent variable models (1984)
