Efficiency Measurement William Greene Stern School of Business New York University

Lab Session 4 Panel Data

Group Size Variables for Unbalanced Panels FarmMilkCowsFarmPrds

Creating a Group Size Variable  Requires an ID variable (such as FARM)  (1) Set the full sample exactly as desired  (2) SETPANEL ; Group = the id variable ; Pds = the name you want limdep to use for the periods variable $ SETPANEL ; Group = farm ; pds = ti $

Application to Spanish Dairy Farms InputUnitsMeanStd. Dev. MinimumMaximum MilkMilk production (liters) 131,108 92,539 14,110727,281 Cows# of milking cows Labor# man-equivalent units LandHectares of land devoted to pasture and crops FeedTotal amount of feedstuffs fed to dairy cows (tons) 57,94147,9813, ,732 N = 247 farms, T = 6 years ( )

Exploring a Panel Data Set: Dairy REGRESS ; Lhs = YIT ; RHS = COBBDGLS ; PANEL $ REGRESS ; Lhs = YIT ; RHS = COBBDGLS ; PANEL ; Het = Group $

Initiating a Panel Data Model

Nonlinear Panel Data Models MODEL NAME ; Lhs = … ; RHS = … ; Panel ; … any other model parts … $ ALL PANEL DATA MODEL COMMANDS ARE THE SAME

Panel Data Frontier Model Commands FRONTIER ; LHS = … [ ; COST ] ; RHS = … [; EFF = …] ; Panel ;... the rest of the model ; any other options $

Pitt and Lee Random Effects FRONTIER ; LHS = … [ ; COST ] ; RHS = … [; EFF = …] ; Panel ; any other options $ This is the default panel model.

Pitt and Lee Model

Pitt and Lee Random Effects with Heteroscedasticity and Time Invariant Inefficiency FRONTIER ; LHS = … [ ; COST ] ; RHS = … [; EFF = …] ; Panel ; HET ; HFU = … ; HFV = … $

Pitt and Lee Random Effects with Heteroscedasticity and Truncation Time Invariant Inefficiency FRONTIER ; LHS = … [ ; COST ] ; RHS = … [; EFF = …] ; Panel ; HET ; HFU = … ; HFV = … ; MODEL = T ; RH2 = One,… $

Pitt and Lee Random Effects with Heteroscedasticity Time Invariant Inefficiency FRONTIER ; LHS = … [ ; COST ] ; RHS = … [; EFF = …] ; Panel ; HET ; HFU = … ; HFV = … $

Schmidt and Sickles Fixed Effects REGRESS ; LHS = … ; RHS = … ; PANEL ; PAR ; FIXED $ CREATE ; AI = ALPHAFE ( id ) $ CALC ; MAXAI = Max(AI) $ CREATE ; UI = MAXAI – AI $ (Use Minimum and AI – MINAI for cost)

True Random Effects Time Varying Inefficiency FRONTIER ; LHS = … [ ; COST ] ; RHS = … $ FRONTIER ; LHS = … [ ; COST ] ; RHS = … ; Panel ; Halton (a good idea) ; PTS = number for the simulations ; RPM ; FCN = ONE (n) ; EFF = … $ Note, first and second FRONTIER commands are identical. This sets up the starting values.

True Fixed Effects Time Varying Inefficiency FRONTIER ; LHS = … [ ; COST ] ; RHS = … $ FRONTIER ; LHS = … [ ; COST ] ; RHS = … ; Panel ; FEM ; EFF = … $ Note, first and second FRONTIER commands are identical. This sets up the starting values.

Battese and Coelli Time Varying Inefficiency FRONTIER ; LHS = … [ ; COST ] ; RHS = … ; Panel ; MODEL = BC ; EFF = … $ This is the default specification, u(i,t) = exp[h(t-T)] |U(i)| To use the extended specification, u(i,t)=exp[d’z(i)] |U(i)| ; Het ; HFU = variables

Other Models There are many other panel models with time varying and time invariant inefficiency, heteroscedasticity, heterogeneity, etc. Latent class, Random parameters Sample selection, And so on….

Lab Session 4 Model Building

Modeling Assignment