Menu
Kelas Ekonomika

Kumpulan Commands Stata




Kali ini admin akan membahas comment yang diperlukan dalam Stata. Admin akan mengelompokkan berdasarkan kategori fungsinya. Simak berikut ini:

Basic Commands

•    help regress

•    display "Hello"

•    di di (l-normal(1.96))*2

•    di sqrt(3.14)

•    describe xl x2

•    des using jeeshim.dbt

•    list in 1/10

•    list male-income

•    list pop* pro? if male==l

•    list-2

•    summarize male grade

•    sum xl-xlOy* post?

•    sum income family if male~=.

•    sum income if (male==l) & (class >=3)

•    tabulate male educate

•    tab male grade, chi2 row col

•    tabi 12 33 \ 34 53, chi2 exact

•    tabi 34 53 23 \ 23 56 34 \ 45 32 21, chi2 all

•    tabstat math english, by(male) stats(n mean sum sd var max min skewness) Management Commands

•    codebook male

•    label data "Pew Internet and American Life Project"

•    label variable male "Gender"

•    label variable male; // to remove a variable lable

•    label define yn 1 yes 0 no

•    label values open yn

•    compare math english

•    cf math english using indiana.dta

•    ci grade if male==0; /* confidence interval */

•    count if math >90

•    lookforgnp

•    notes var: Need to be verified

•    update all

•    net search Spost

•    net from 

•    net describe spost9_ado

•    net spost9_ado

•    net get spost9_do

•    sscwhatsnew

•    ssc describe

•    ssc install

Operators
• + _ * / A
•    >, >=, <, <=, = (equal), ~= (not equal)
•    & (and), | (or), ~ (not)
   +=, =+, -=, =-, /=, %=, &=, |=, *=
•    in (in if command)
•    + (other variables), 1/b (from a through b),. (missing values)
•    wild card (*, ?, /, -)
•    concatenation (+)
Math Functions

•    abs(); sin(); cos(); tan();asin(); acos(); atan()
•    ceil(); floor(); int() or trunc(); round()
•    exp(); sqrt(); log(); ln(); logl0()
•    min(); max(); sign(); sum(); mod(x,y); comb(x,k)
Probability Distribution
•    binomal(h,k,p) //joint cumulative distribution of bivariate normal
•    chi2(df,x) // cumulative chi squared distribution
•    chi2tail(n,x) // reverse of chi2()
•    F(dfl, df2, f) // cumulative F distribution
•    Ftail(dfl, df2, f) // reverse cumulative (upper-tail) F
•    normal(z); normal(1.96) // returns .9750002
•    ttail(df, t) // reverse cumulative (upper-tail) T
•    uniform() // uniform distributionreverse cumulative (upper-tail) T
•    di chi2tail(l0,18.31) // returns .04995417, p-value
•    di F(5, 10, 3.325) // returns .9499661
•    di Ftail(5, 10, 3.325) // returns .05000, the pa-value
•    di (l-normal(z))*2 // compute the p-value for the two-tailed test
•    di ttail(20, 2.086) // returns .02499818
•    di ttail(df, t)*2 // compute the p-value for the two-tailed test
String Functions


•    char(n); length(s); trim(s); ltrim(s); rtrim(s)
•    string(n); substr(s,begin, length)
•    real(s); reverse(s); wordQ; lowerQ; upper()

Handling Data Sets

•    use "k:\kucc625\open.dta", clear
•    use j eeshim. dta, clear nolabel
•    use using jeeshim.dta if gender==l, clear nolabel
•    save "c:\kucc625\open.dta", replace
•    save open.dta, replace nolabel
•    saveold "c:\kucc625\jeeshim.dta", replace
•    log using "c:\kucc625\open.log", append
•    log using open.log, append text
•    log on // log off; log close
•    logcmd using "c:\kucc625\open_cmd.log", replace
•    logcmd on // logcmd off; logcmd close
Import
•    infile a b c usingjeeshim.txt, clear
•    infile strl5 name float weight int height using student.txt
•    inf id _skip(l) ql-q3 using student.txt, clear
•    inf str20 id long (ql-q3) using student.txt, clear
•    inf id double (ql0-ql3) income if income >50000 using studenttxt
•    inf using student.dct, clear
•    infix year 1-4 gnp 5-9 interest 10-13 using macro.txt
•    infix using macro.dct in 1/100, clear
•    insheet using jeeshim.csv, clear
•    insheet a b c d e f g using student.txt, comma clear
•    insheet using student.txt, delim("#") clear
Export
•    odbc list
•    odbc load ID=year gnp interest in 1/500, table("macro") dsn("j eeshim")
•    outfileusingjeeshim.txt
•    outfile x 1 -x 10 using j eeshim.txt, wide replace
•    outfile using jeeshim.txt, nolabel noquote replace
•    outsheet using jeeshim.xls, nolabel /* tab delimited */
•    outsheet using jeeshim.xls, comma replace

Editing
•    keep gender grade korean math english if gender=l
•    keepidql-q20
•    drop templ-temp5 if gender=0
•    drop temp* pro? if income <5000
•    drop in 1/10
•    drop if gender==l in 1/100
•    edit
•    editinl0/-5
•    edit if gender==0
•    edit in 1/100 if income >5000
•    edit male class if income >=30000
•    mark ynmiss
•    markout yn_miss ql-qlO //0 if any one of variable has missing
•    isid college // to check for unique identifiers Recoding
•    generate gender; gen gender=male
•    gen square=gnpA2
•    gen grade=(score <= 90 | attendance=0) if final~=.
•    egen avg = rowmean(english math stats)
•    egen gnpbar = mean(gnp), by(country)
•    replace gender=0
•    replace gender=l if male=l
•    replace male=l in 3
•    recode class 1=0 2=1 *=.
•    recode class 1/3=0 4=1 if male==0
•    recode grade 1 2 3 5=1 4=2
•    recode grade 9999=.
•    recode grade min/5=min
•    recode grade 6/max=max
Reshaping Data Sets
•    set obs 100 // to change the numnber of observations
•    sort male grade
•    gsort -grade name, gen(rank)
•    append using c:\data\class
•    app using c:\data\class, keep(id state ql-qlO)
•    expand 5 in -10/-1 // duplicate observations n-1 times
•    merge using school // one-to-one merging
•    merge state using school // match merging
•    merge state using school university, update replace
•    joinby id using secondary, unmatched(master) // unm(both), unm(using)
•    move male grade
•    order grade male // order variables as listed
•    rename male gender; /* from male to gender */
•    expand 5 if state=="IN" // duplicate a subset of observations
•    collapse a b (sd) c (count) d (max)
•    collapse a b (sd) c (count) d (max), by state
•    contract gender degree area, freq(count) zero
•    reshape long choice, i(id) j (orders)
•    stack bestl-best3, into(best) clear
•    pkshape id row coll-col3, order(abc cab bca) outcome(y) sequence(rows) treat(treat) period(columns)
•    compress // all variable
•    compress name grade
•    xpose, clear vamame

Ordinary Least Squares (OLS)
•    regress dv ivl iv2
•    regress depend indep 1 -indep 10, noconstant
•    regress income school job location if gender==0
•    regress income school job location if gender==0, noconstant level(95)
•    predict p // xb
•    predict r, residual
•    fitstat
•    quietly fitstat, saving(modell)
•    fitstat, using(modell)
•    fitstat, dif
•    bgodfrey, lag(l 2 3); estat bgodfrey, lag(l 2 3)
•    dwstat; estat dwatson, lag(l 2 3)
•    stepwise, pr(.2): regress y xl-xlO // backward stepwise regression
•    constraint define 1 dl+d2+d3=0 // LSDV2 in Stata
•    constraint define 2 gl+g2+g3+g4=0
•    cnsregs y xl x2 dl-d3 gl-g4, constraint^ 2)
Hypothesis Test


•    test school /* Wald Test */
•    test school location; test school=location
•    test job; test school, accumulate
•    lrtest, saving(O); /* Likelihood Ratio Test */
•    lrtest, saving(l)
•    lrtest, using(l) model(O) 
•    boxcox //Box-Cox regression model
•    eivreg // errors-in-variables regression
•    ffacpoly // Fractional polynomial regression
•    frontier //Stochastic frontier models
•    glm // generalized linear model
•    intreg //interval regression
•    ivreg //instrumental variables (two-stage least squares) regression
•    ivreg dv ivl iv2 (iv3= xl x2 x3) iv4 iv5
•    mfp //multivariable fractional polynomial models
•    mvreg //multivariate regression
•    newey //Regression with Newey-West standard errors
•    nl //nonlinear least-squares estimation
•    orthog //Orthogonalize variables and compute orthogonal polynomials
•    prais dvl rhs, rho(tscorr) twostep //Prais-Winsten two-step
•    prais dvl rhs, rho(dw) // iterative two-step
•    prais dvl rhs, rho(dw) core // Cochrane-Orcutt
•    qreg //Quantile (including median) regression
•    reg3 //three-stage estimation for systems of simultaneous equations
•    reg3 (dvl xl x2) (dv2 xl x3)
•    reg3 (dvl dv2 = xl x2 x3)
•    reg3 (dvl dv2 = xl x2 x3) (dv3 xl x3)
•    rocfit//fit ORC model
•    rreg//robust regression
•    stcox //fit Cox proportional hazards model
•    streg //fit parametric survival model
•    sureg //Zellner's seemingly unrelated regression
•    stepwise //stepwise estimation
•    treatreg //treatment-effects model
•    treatreg y xl x2 x3, treat(x4=zl z2) twostep
•    vwls //variance-weighted least squares

Panel data

•    tsset group year // set group and time
•    xtreg y xl x2, re i(year) // random effect model
•    xtreg y xl x2, fe i(group) // random effect model
•    xtreg y xl x2, be i(group) // between effect model
•    areg // linear regression with a large dummy-variable set
•    xtabond // Arellano-Bond linear, dynamic panel-data estimator
•    xtcloglog // Random-effects, population-averaged cloglog models
•    xtgee // fit population-averaged panel-data models using GEE
•    xtfrontier // stochastic frontier models for panel data
•    xtgls // fit panel-data models using GLS
•    xthtaylor // Hausman-Taylor estimator for error components models
•    xtinreg // random-effects interval data regression models
•    xtivreg // Instrumental variables and two-stage least squares
•    xtlogit //fixed-effects, random-effects, population-averaged logit
•    xtmixed // multilevel mixed-effects linear regression
•    xtprobit // random-effects and population averaged probit models
•    xttobit // random-effects tobit models
•    xtnbreg //fixed-effects, random-effects, and population-averaged NB
•    xtpcse // Prais-Winsten models with panel-corrected standard errors
•    xtpoisson //fixed-effects, random-effects, population-averaged Poisson
•    xtrc // random-coefficients models
•    xtregar // fixed-and random-effects linear models with an AR(1)

Binary Logit/Probit
•    logit dv ivl iv2
•    logit card income school job if gender==0, nolog nocon
•    logistict dv ivl iv2
•    probit dv ivl iv2
•    predict p
•    prchange; prchange income, x(school=l job=l)
•    prchange school, x(income=10000) help
•    prtab income school, rest(mean)
•    prgen income, ffom(O) to(10000) x(school=l) rest(mean)
•    prvalue, rest (mean)
•    prvalue, x(income=10000 j ob= 1) rest(mean)
Ordinal and Multinomial
•    ologit dv ivl iv2 // Ordinal
•    ologit grade income school job if gender==0, nolog nocon
•    oprobit dv ivl iv2
•    omodel logit card income school job // Approximate LR test
•    mlogit dv ivl iv2; /* Nominal */
•    mlogit mode income school job, basecategory(l) nolog
•    mlogtest, lr
•    mlogtest, wald
•    mlogtest, hausman base
•    mprobit // multinomial probit regression
•    clogit dv ivl iv2, group(var) // Conditional logit
•    clogit mode income school job, group(gender) nolog
•    nlogit // nested logit regression
Special Logit/Probit
•    asmprobit // alternative-specific multinomial probit
•    biprobit (dvl=rhsl) (dv2=rhs2) // bivariate probit
•    glogit //logit and porbit for grouped data
•    heckprob dv rhs, select(rhs2) // probit model with selection
•    hetprob // heteroskedastic probit
•    ivprobit // probit model with endogenous regressions
•    rologit // rank-ordered logistic
•    scobit // Skewed logit
•    slogit // sterotype logistic
•    xtlogit // logit models for panel data
•    xtprobit // probit models for panel data
Event Count Data Models
•    poisson dv ivl iv2
•    nbreg dv ivl iv2
•    zip dv ivl iv2 // zero-inflated Poisson Model
•    zinb dv ivl iv2 // zero-inflated NB Model
•    ztp dv ivl iv2 // zero-truncated Poisson Model
•    ztnb dv ivl iv2 // zero-truncated NB Model
Truncated/Censored/Self-selected
•    cnreg // Censored-normal regression
•    heckman // Heckman selection model
•    ivtobit // Tobit model with endogenous regressors
•    tobit // Tobit regression
•    truncreg // truncated regression
•    ztp dv ivl iv2 // zero-truncated Poisson Model
•    ztnb dv ivl iv2 // zero-truncated NB Model
Related Commands
•    bootstrap // bootstrap sampling and estimation
•    bsample // Sampling with replacement
•    jackknife // Jackknife estimation
•    impute // imputation
•    permute // Monte Carlo permutation test
•    simulate // Monte Carlo simulation
•    sampsi // sample size and power determination
T-Test
•    ttest grade==10
•    ttest grade, by(male)
•    ttest grade, by(male), unequal welch
•    ttest math=english; ttest math==english, unpaired
•    ttesti 100 88.1 5.2 90; /* N mean sd hypothesis */
•    ttesti 100 88.1 5.2 200 91 10.2; /* N1 meanl sdl N2 mean2 sd2 */
•    ttesti 100 88.1 5.2 200 91 10.2, unequal
•    mean // estimate means
•    total // estimate totals
•    ratio // estimate ratios
•    proportion // one- and two-sample tests of proportions
•    ci // confidence intervals for means, proportions, and counts
ANOVA
•    anova score gender
•    anova score gender year gender*year
•    oneway score gender, tabulate
•    loneway //large one-way ANOVA, random effect, and reliability
•    sdtest // Variance-comparison test
•    Related: .manova; .pkshape; .xtmixed
Correlation Analysis


•    correlate gnp interest inflation
•    corr gnp interest inflation, covariance
•    pcorr xl-xlO // partial correlation coefficients
•    pwcorr gnp interest inflation, sig
•    pwcorr gnp interest inflation, print(5) // .05 significance level
•    pwcorr gnp interest inflation, sig star(.05) // .05 level
Factor Analysis
•    factor xl-x30 // by default pcf (principal component factor)
•    factor xl-x30, ml // maximum likelihood factor
•    factor xl-x30, factors(5)
•    rotate, varimax // orthogonal, oblique, quartimax, equamax, parsimax, promax
•    pea // principal component analysis
Other Analysis
•    alphar // Cronbach's alpha
•    ca // correspondence analysis
•    canon // Canonical correlation
•    cluster // cluster analysis
•    mvreg // multivariate regression
•    manova // multivariate MANOVA
•    mds // multidimensional scaling for two way data
•    mdslong
•    mdsmat
•    biplot
NONPARAMETRIC ANALYSIS
•    swilk math english
•    sffanciaxl-xlO
•    ranksum //Equality tests on unmatched data
•    signrank math=english // Equality tests on matched data
•    runtest // test for random order
•    spearman xl-x 10
•    kwallis score, by(gender)
•    ksmimov math, by(area)
•    alpha xl-xlO, item
•    kappa evall eval2
•    bitest //Binomial probability test
•    prtest // one- and two-sample tests of proportions
Graphics Basics
•    sysuse auto, clear
•    graph bar (mean) mpg turn, by(foreign)
•    graph bar (mean) mpg turn, over(foreign)
•    graph hbar (mean) mpg turn, over(foreign)
•    graph hbar (mean) mpg, over(foreign) over(class)
•    graph hbar (mean) mpg, over(class) over(foreign)
•    graph hbar (mean) mpg, over(class) over(foreign, sort(l) descending)
•    graph hbar (sum) mpg turn, over(class) stack
•    graph hbar (sum) mpg turn, over(class) by(foreign)
•    graph hbar (sum) mpg turn, over(class) by(foreign) stack
•    graph dot (mean) mpg, over(class)
•    graph dot (mean) mpg, over(class) over(foreign)
•    graph matrix mpg price turn, half
•    graph pie, over(class)
•    graph pie mpg turn trunk, plabel(_all name)
Scatter and Two-way Plotting
•    scatter mpg weight
•    scatter mpg weight, sort
•    scatter mpg weight, sort connect(l)
•    scatter mpg weight, sort title("MPG versus Weight") subtitle("Year 2006")
•    scatter mpg weight, title("MPG versus Weight") caption(" Source: Stata Corp. 2006")
•    scatter mpg weight, title("MPG versus Weight") xsize(4) ysize(3)
•    scatter mpg weight, ytitle("MPG (Mileage)") xtitle("Car Weight")
•    scatter mpg weight, title("MPG versus Weight") ylabel(#8) xlabel(0(2000)6000)
•    scatter mpg weight, title("MPG versus Weight") ylabel(minmax) xlabel(minmax)
•    scatter mpg weight, title("MPG versus Weight") yscale(log) xlabel(#5) // log scales
•    scatter mpg weight, sort xline(4000) yline(25)
•    scatter mpg weight, title("MPG versus Weight") msymbol(triangle)
•    scatter mpg weight || fpfit mpg weight
•    twoway fpfitci mpg weight
•    twoway fpfitci mpg weight || scatter mpg weight, m(d)
•    scatter mpg weight, sort title("MPG versus Weight") m(diamond) by(foreign)
•    scatter mpg weight, sort m(t) by (foreign, total row(l))
•    twoway fpfitci mpg weight, sort m(t) by(foreign, total row(l))
•    twoway fpfitci mpg weight || scatter mpg weight, sort m(t) by (foreign, total row(l))
•    scatter mpg turn weight
•    scatter mpg turn weight, yline(30) xline(3500)
•    scatter mpg trunk turn weight
•    scatter mpg weight || scatter trunk weight || scatter turn weight
•    scatter mpg weight, sort c(l) || line trunk weight, sort || scatter turn weight
•    twoway (line mpg weight, sort c(l)) (dropline trunk weight, sort) (scatter turn weight)
Plotting by Functions

•    twoway function y=xA3, range(-5 5) xsize(4) ysize(3) xlabel(#10) xline(0)
•    twoway function y=normalden(x), range(-5 5) xsize(4) ysize(2) xlabel(#10) xline(0)
•    twoway function y=l/sqrt(2*_pi)*exp(-xA2/2), range(-5 5) xsize(4) ysize(2) xlabel(#10) xline(0)
•    twoway function y=normalden(x), range(-4 -1.96) xlabel(#10) xline(0) recast(area) || function y=normalden(x), range(1.96 4) recast(area) || function y=normalden(x), range(-1.96 1.96) lstyle(foreground)
•    twoway function t=tden(3, x), range(-5 5) xsize(4) ysize(2) xline(O)
•    twoway function t=tden(l, x), range(-5 5) xsize(4) ysize(2) color(blue) lstyle(plsolid) xlabel(-5(l)5) recast(area) || function z=normden(x), range(-5 5) color(maroon) lwidth(thick)
•    scatter gear ratio headroom, xsize(4) || function y=x, range(0 5)
•    twoway function c=chi2(l,x), range(0 5) xsize(4) ysize(3) yline(.5)
•    twoway function c=Fden(5, 10, x), range(0 5) xsize(4) ysize(3) yline(.3)