Linear Probability, Logit, and Probit Models (Quantitative Applications in the Social Sciences)

Ordinary regression analysis is not appropriate for investigating dichotomous or otherwise "limited" dependent variables, but this volume examines three techniques -- linear probability, probit, and logit models -- which are well-suited for such data. It reviews the linear probability model and discusses alternative specifications of non-linear models. Using detailed examples, Aldrich and Nelson point out the differences among linear, logit, and probit models, and explain the assumptions associated with each. Local Disk......Page 0 cover......Page 1 cover-0......Page 2 cover-1......Page 3 page_1......Page 4 page_2......Page 5 page_3......Page 6 page_4......Page 8 page_5......Page 10 page_6......Page 11 page_7......Page 12 page_9......Page 13 page_10......Page 14 page_11......Page 15 page_12......Page 16 page_13......Page 17 page_14......Page 18 page_15......Page 19 page_16......Page 20 page_17......Page 22 page_18......Page 24 page_19......Page 25 page_20......Page 26 page_21......Page 27 page_22......Page 28 page_23......Page 29 page_24......Page 30 page_25......Page 31 page_26......Page 32 page_27......Page 33 page_28......Page 34 page_29......Page 35 page_30......Page 36 page_31......Page 37 page_32......Page 38 page_33......Page 39 page_34......Page 40 page_35......Page 41 page_36......Page 42 page_37......Page 43 page_38......Page 44 page_39......Page 45 page_40......Page 46 page_41......Page 47 page_42......Page 48 page_43......Page 49 page_44......Page 50 page_45......Page 51 page_46......Page 52 page_47......Page 53 page_48......Page 54 page_49......Page 55 page_50......Page 56 page_51......Page 57 page_52......Page 58 page_53......Page 59 page_54......Page 60 page_55......Page 61 page_56......Page 62 page_57......Page 63 page_58......Page 64 page_59......Page 65 page_60......Page 66 page_61......Page 67 page_62......Page 68 page_63......Page 70 page_64......Page 72 page_65......Page 74 page_66......Page 75 page_67......Page 76 page_68......Page 77 page_69......Page 78 page_70......Page 79 page_71......Page 80 page_72......Page 81 page_73......Page 83 page_74......Page 84 page_75......Page 85 page_76......Page 86 page_77......Page 88 page_78......Page 89 page_79......Page 90 page_80......Page 91 page_81......Page 92 page_82......Page 93 page_83......Page 94 page_85......Page 95 page_86......Page 96 page_87......Page 98 page_88......Page 99 page_89......Page 100 page_90......Page 101 page_91......Page 102 page_93......Page 103 page_94......Page 105 page_95......Page 106 Funded By Dsu Title Iii 2007-2012. The Linear Probability Model -- Specification Of Nonlinear Probability Models -- Estimation Of Probit And Logit Models For Dichotomous Dependent Variables -- Minimum Chi-square Estimation And Polytomous Models -- Summary And Extensions. John H. Aldrich, Forrest D. Nelson. Bibliography: P. 93-94. After showing why ordinary regression analysis is not appropriate for investigating dichotomous or otherwise 'limited' dependent variables, this volume examines three techniques which are well suited for such data. It reviews the linear probability model and discusses alternative specifications of non-linear models