统计学软件考核

《统计学软件考核》由会员分享，可在线阅读，更多相关《统计学软件考核（9页珍藏版）》请在装配图网上搜索。

1、统计软件的实践考核题目1. 全国 31 个省、直辖市、自治区分别在铁路运输业、道路运输业、水上运输业、 航空运输业等领域的就业人数数据，请见珂 请作出该就业数据的轮廓 图与星图，并完成其数据特征分析与相关分析。解：录入数据(文件名：yunshu.csv)rt rt rtXtieludaolushuishanghangkongguandaozhuangxiecangchuyouzheng1199577273618324666535947510561288782206222129851413211928236235132721933084793351533138157255683306567117

2、421171133311441149278249274519642324118420224495510825973428314076NA1644630423101661190401423783922319150383917867116832340477687716098983153791117138412215209018813914376624310367337674177200383020099452222027985053960843145467921304093240010102333623959778329133681038136773178616441911112810515972

3、42648010887106209301240055066121242541112095146134116NA532790843227513134030910527114927152341229352553632842141461574106077789448448083030663721578151582442234278704051411335472568822729532031616120139 2342785768779574109292324833954171783833145880157579137788110101132554184181876901103424476174631

4、6982735077392731919274523804856010610975531954503334151670412020661088265910430522972143616309272512121503419638507614198304615105347602222285351658981488410384NA7192440029627232360521195737109153532861893657339740032424317975479511426018NA389235481225725253766272439716185253721770431081744226266746

5、60NA372NANA23919012727659921173122101222826613410951139057282851684505234616425110574487171392929222591881841277NA2471129383330301696415857801710NA450649471131315535784780NA1120233472928293414328一、缺失值的处理由于数据中有较多的缺失值，所以我们要编写程序有均值将缺失值替换，具体步 骤如下：1、判断每行有没有缺失值，再找出缺失值的行数，进行完整的观察数据的个 数；2、给每一列求均值，将数据框转换成矩阵，

6、对矩阵求均值；3、将rt数据里面的缺失值赋值给均值，将数据框的变量链接到内存里面，以 便调用，将每一个变量中的缺失值转换成它的均值。以下是程序代码：complete.cases(rt)which(complete.cases(rt)=F) sum(complete.cases(rt) col_Means-colMeans(rt,na.rm=TRUE) aa=as.matrix(rt) mean_aa-mean(aa)mean_aa rtis.na(rt)-mean_aa attach(rt) tieluis.na(tielu)-col_Means2 daoluis.na(daolu)-col_

7、Means3 shuishangis.na(shuishang)-col_Means4 hangkongis.na(hangkong)-col_Means5 guandaois.na(guandao)-col_Means6 zhuangxieis.na(zhuangxie)-col_Means7 cangchuis.na(cangchu)-col_Means8 youzhengis.na(youzheng)-col_Means9 rt$tielu-tielurt$guandao-guandao rt$daolu-daolu rt$shuishang-shuishang rt$hangkong-

8、hangkong rt$zhuangxie-zhuangxie rt$youzheng-youzheng rt$cangchu rt$tielu rt$guandao rt$daolu rt$shuishang rt$hangkong rt$zhuangxie rt$youzheng rt$cangchu rtX tielu daolu shuishang hangkong guandao zhuangxie cangchu youzheng1199577273618324.0066653594751056.00128878220622212985141321192.0082362351327

9、2.00193308479335153313815725568.00330656711742.001171133311441149278249274.0051964232411.008420224495 5 1082597342831.00407615711644.006304231016611904014237839223.0019150383917867.001168323404776877160989831.00537911171384.00122152090188139143766243103.0067337674177.002003830200994522220279850539.0

10、060843145467921.00304093240010 102333623959778329.00133681038136773.00178616441911 112810515972426480.001088710620930.00124005506612 124254111209514613.00411615715327.0090843227513 134030910527114927.00152341229352.0055363284214 14615741060777894.0048448083030.0066372157815 158244223427870405.001411

11、3354725688.00227295320316 161201392342785768.0077957410929.00232483395417 178383314588015757.00913778811010.00113255418418 18769011034244761.0074631698273.0050773927319 192745238048560106.0010975531954503.003341516704120 20661088265910430.0052297214361.0063092725121 215034196385076.0014198304615.001

12、053476022 222853516589814884.001038415717192.0044002962723 236052119573710915.00353286189365.0073397400324 2431797547951142.00601815713892.0035481225725 253766272439716.001852537217704.0031081744226 2667466016669.93372157114750.33239190127 2765992 117312210.001222826613410.0095113905728 285168450523

13、46.001642511057.0044871713929 2922259188184.0012771571247.001129383330 30169641585780.0017101571450.00649471131 31553578478016669.931120233472928.00293414328二、画轮廓图读出数据之后，下面编写轮廓图函数outline()函数 outline-function(x,txt=TRUE) if(is.data.frame(x)=TRUE)x-as.matrix(x) m-nrow(x);n-ncol(x) plot(c(1,n),c(min(x)

14、,max(x),type=n, main=The outline graph of Data, xlab=Numer,ylab=Value)for(i in 1:m) lines(xi,col=i) if(txt=TRUE) kNumer上述即为所给数据由R软件画出的轮廓图。由轮廓图可以直接的看出，全 国 31 个省、直辖市、自治区分别在铁路运输业、道路运输业、水上运输业、 航空运输业等的分散情况。主要还是铁路运相对人数最高；其次是输邮政运输 业，管道运输业就业人数最低。三、画星图画星图步骤及图如下：stars(rt,full=FALSE,draw.segment=TRUE,key.loc=c

15、(8,0.5),mar=c(2,0,0,0)guan, zhuanqxi cangcnu youzheng juielu由上面的星图可以看出：（1）以北京市为例。在铁路运输业、道路运输业、水上运输业、航空运输业等 领域的就业人数中“装卸搬运和运输代理业运输业”就业人数相比其他几个领域 的高，管道运输也就业人数相对最少。（2）对全国 31 个地区运输业领域就业人数进行比较，可以看出在这 31 个地区 中北京、上海、江苏、广东这四个地区的运输业就业人数在全国占有重要比重。 （3）其余的以西藏为例,由于运输业领域的就业人数少在图中几乎看不到。 相关性检验如下：读数据并计算协方差矩阵和相关矩阵mean

16、（rt）cov（rt） cov(rt)XtieludaolushuishangX82.66667-151539.9-226976.1-44349.96tielu -151539.86667 1327472132.3 645665149.8 -125595145.80 daolu -226976.13333 645665149.8 7408150286.2 1171478317.66shuishang -44349.95747 -125595145.8 1171478317.7 461112531.33 hangkong-28978.16667-27395813.3 1503748080.220

17、8328679.54guandao-3029.166671901397.9 62921906.522692716.98zhuangxie -49860.25556 -55856905.6 1056565043.0 238445527.63 cangchu-37653.8333383212970.6 549363541.0119407330.07youzheng-49407.8000099207948.8 2414127453.9343413584.66hangkongguandaozhuangxiecangchuyouzhengX-28978.17-3029.167-49860.26-3765

18、3.83-49407.8tielu-27395813.341901397.900-55856905.5683212970.6099207948.8daolu1503748080.1662921906.5001056565043.00549363540.992414127453.9shuishang208328679.54 22692716.984238445527.63119407330.07343413584.7hangkong536155687.286017983.933318789647.87121562614.39599331556.4guandao6017983.934531905.

19、40011980629.372505625.4012464247.8zhuangxie318789647.87 11980629.367290001139.8298368816.38335769353.6cangchu121562614.392505625.40098368816.3873561993.70169035228.4youzheng599331556.4112464247.800335769353.59169035228.451010190169.2四、作相关性检验 cor.test(tielu+guandao,data=rt) cor.test(tielu+guandao,dat

20、a=rt)Pearsons product-moment correlationdata: tielu and guandaot = 0.1321, df = 29, p-value = 0.8959alternative hypothesis: true correlation is not equal to 095 percent confidence interval:-0.3327159 0.3755919sample estimates:cor0.02451432同理用同样的方式分别做 tielu,guandao ,daolu 等的相关性检验，重复上面 的检验可以得出， tielu,daolu,gua ndao,ha ngkon g,ca ngchu,youzhe ng,shuisha ng,zhua ngxie，两两不相关； 也就是说每一种行业就业之间是没有影响的。五、总结由以上分析可以分析出：北京、上海、江苏、山东、河南、广东的就业情况 比较好，其余省份的就业情况都比较差，从星图可以明显看出在比较发达的省份 在铁路运输，仓储，道路，航空运输行业的就业情况远比内陆其他不发达城市的 就业情况好得多，而内陆不发达省份就业主要以邮政、装卸、管道就业为主。