NOTE: If you want to see the design effect or the misspecification effect, use estat effects after the command.
Page 231, table 7.1
use https://stats.idre.ucla.edu/stat/stata/examples/lohr/htpop.dta, clear
tabulate height
height | Freq. Percent Cum.
------------+-----------------------------------
136 | 1 0.05 0.05
140 | 1 0.05 0.10
141 | 2 0.10 0.20
142 | 1 0.05 0.25
143 | 6 0.30 0.55
144 | 3 0.15 0.70
145 | 4 0.20 0.90
146 | 3 0.15 1.05
147 | 14 0.70 1.75
148 | 11 0.55 2.30
149 | 13 0.65 2.95
150 | 20 1.00 3.95
151 | 15 0.75 4.70
152 | 18 0.90 5.60
153 | 28 1.40 7.00
154 | 38 1.90 8.90
155 | 38 1.90 10.80
156 | 57 2.85 13.65
157 | 53 2.65 16.30
158 | 49 2.45 18.75
159 | 55 2.75 21.50
160 | 77 3.85 25.35
161 | 72 3.60 28.95
162 | 66 3.30 32.25
163 | 62 3.10 35.35
164 | 61 3.05 38.40
165 | 60 3.00 41.40
166 | 75 3.75 45.15
167 | 79 3.95 49.10
168 | 62 3.10 52.20
169 | 79 3.95 56.15
170 | 72 3.60 59.75
171 | 56 2.80 62.55
172 | 57 2.85 65.40
173 | 45 2.25 67.65
174 | 52 2.60 70.25
175 | 57 2.85 73.10
176 | 49 2.45 75.55
177 | 54 2.70 78.25
178 | 57 2.85 81.10
179 | 40 2.00 83.10
180 | 35 1.75 84.85
181 | 43 2.15 87.00
182 | 29 1.45 88.45
183 | 26 1.30 89.75
184 | 29 1.45 91.20
185 | 23 1.15 92.35
186 | 21 1.05 93.40
187 | 19 0.95 94.35
188 | 17 0.85 95.20
189 | 15 0.75 95.95
190 | 10 0.50 96.45
191 | 14 0.70 97.15
192 | 10 0.50 97.65
193 | 9 0.45 98.10
194 | 7 0.35 98.45
195 | 2 0.10 98.55
196 | 7 0.35 98.90
197 | 8 0.40 99.30
198 | 4 0.20 99.50
199 | 2 0.10 99.60
200 | 4 0.20 99.80
201 | 1 0.05 99.85
204 | 1 0.05 99.90
206 | 2 0.10 100.00
------------+-----------------------------------
Total | 2,000 100.00
Page 231 Figure 7.1
cumul height, gen(ch)
sort height
graph scatter ch height, con(stepstair) ylabel(0 .4 .8, nogrid angle(0)) ///
ytitle("F(y)") xtitle("Height Value, y") xlabel(140(20)200) msym(i)
Page 232 Figure 7.2
histogram height, discrete xtitle("Height Value, y") ylabel( , nogrid)
Page 232 Figure 7.3
use https://stats.idre.ucla.edu/stat/stata/examples/lohr/htsrs.dta, clear histogram height, bin(24) ylabel( , nogrid)
Page 232, figure 7.4
use https://stats.idre.ucla.edu/stat/stata/examples/lohr/htstrat.dta, clear histogram height, bin(21) ylabel( , nogrid angle(0))
Page 233, figure 7.5
NOTE: The top figure has been skipped for now.
cumul height, gen(ch)
sort height
graph scatter ch height, con(stepstair) ylabel(0 .4 .8, nogrid angle(0)) ///
ytick(.2 .6) xlabel(140(10)200) xtitle("Height Value, y") msym(i)
Page 234, table 7.2
NOTE: We could only get the mean for the last column.
Population:
use https://stats.idre.ucla.edu/stat/stata/examples/lohr/htpop.dta, clear
tabstat height, s(mean median p25 p90 var)
variable | mean p50 p25 p90 variance
-------------+--------------------------------------------------
height | 168.616 168 160 184 124.5038
----------------------------------------------------------------
SRS:
use https://stats.idre.ucla.edu/stat/stata/examples/lohr/htsrs.dta, clear
tabstat height, s(mean median p25 p90 var)
variable | mean p50 p25 p90 variance
-------------+--------------------------------------------------
height | 168.94 169 160 184 122.6396
----------------------------------------------------------------
Stratified, no weights:
use https://stats.idre.ucla.edu/stat/stata/examples/lohr/htstrat.dta, clear
tabstat height, s(mean median p25 p90 var)
variable | mean p50 p25 p90 variance
-------------+--------------------------------------------------
height | 164.645 163 157 178 93.40601
----------------------------------------------------------------
Stratified, with weights:
use https://stats.idre.ucla.edu/stat/stata/examples/lohr/htstrat.dta, clear gen wt = 6.25 replace wt = 25 if gender == "M"
svyset [pweight = wt]
pweight: wt
VCE: linearized
Strata 1: <one>
SU 1: <observations>
FPC 1: <zero>
svy: mean height
(running mean on estimation sample)
Survey: Mean estimation
Number of strata = 1 Number of obs = 200
Number of PSUs = 200 Population size = 2000
Design df = 199
--------------------------------------------------------------
| Linearized
| Mean Std. Err. [95% Conf. Interval]
-------------+------------------------------------------------
height | 169.0156 .9766307 167.0898 170.9415
--------------------------------------------------------------
Page 236, figure 7.6
use https://stats.idre.ucla.edu/stat/stata/examples/lohr/syc.dta, clear histogram age, freq ylabel( , nogrid angle(0))
Page 237 Figure 7.7
NOTE: This has been skipped for now.
Page 237, figure 7.8
graph box age [pweight = finalwt], over(stratum) ytitle(Age) /// ylabel(12(2)24, nogrid angle(0)) b1title(Stratum)
Page 238, figure 7.9
sort stratum age by stratum age: egen sumwt = total(finalwt) graph scatter age stratum [pweight = sumwt], msymbol(Oh) msize(vsmall) /// ylabel(12(2)24, nogrid angle(0))
Page 238, figure 7.10
graph box age if stratum == 5 [pweight = finalwt], over(psu) /// ylabel( , nogrid angle(0)) ytitle(Age) b1title(psu Number) /// medline(lwidth(thick) lcolor(red)) medtype(cline) cwhiskers /// lines(lpattern(dash)) capsize(20)
Page 239 Figure 7.11
sort psu age by psu age: egen wt711 = total(finalwt) gen fake_psu = psu recode fake_psu (17=1) (19=2) (39=3) (41=4) (42=5) (48=6) (49=7) list psu fake_psu if stratum == 5 graph scatter age fake_psu if stratum == 5 [pweight = wt711], /// msymbol(Oh) msize(vsmall) ylabel( , nogrid angle(0)) /// xlabel(1 "17" 2 "19" 3 "39" 4 "41" 5 "42" 6 "48" 7 "49") xtitle(psu Number)
Page 247 Figure 7.12
NOTE: This has been skipped for now.