《Matlab回归分析.doc》由会员分享，可在线阅读，更多相关《Matlab回归分析.doc（7页珍藏版）》请在三一办公上搜索。
1、1、 考察温度*对产量y的影响，测得以下10组数据：温度20253035404550556065产量kg13.215.116.417.117.918.719.621.222.524.3求y关于*的线性回归方程，检验回归效果是否显著，并预测*=42时产量的估值及预测区间置信度95%.*=20:5:65;Y=13.2 15.1 16.4 17.1 17.9 18.7 19.6 21.2 22.5 24.3;*=ones(10,1) *;plot(*,Y,r*);b,bint,r,rint,stats=regress(Y,*); b,bint,stats; rcoplot(r,rint) %残差分析
2、，作残差图结果：b = 9.1212 0.2230bint = 8.0211 10.2214 0.1985 0.2476stats =0.9821 439.8311 0.0000 0.2333即；的置信区间为的置信区间为; =0.9821 , F=439.831,p=0.0000 ,p0.05, 可知回归模型 y=9.1212+0.2230* 成立.将*=42带入得到18.4872.从残差图可以看出，所有数据的残差离零点均较近，且残差的置信区间均包含零点，这说明回归模型y=9.1212+0.2230*能较好的符合原始数据。2*零件上有一段曲线，为了在程序控制机床上加工这一零件，需要求这段曲线的
3、解析表达式，在曲线横坐标*i处测得纵坐标yi共11对数据如下：求这段曲线的纵坐标y关于横坐标*的二次多项式回归方程。t=0:2:20;s=0.6 2.0 4.4 7.5 11.8 17.1 23.3 31.2 39.6 49.7 61.7;T=ones(11,1) ,t,(t.2);b,bint,r,rint,stats=regress(s,T);b,stats;Y=polyconf(p,t,S)plot(t,s,k+,t,Y,r) %预测及作图b = 1.0105 0.1971 0.1403stats = 1.0e+04 * 0.0001 1.3773 0.0000 0.0000图形为：3混
4、凝土的抗压强度随养护时间的延长而增加，现将一批混凝土作成12个试块，记录了养护日期*日及抗压强度ykg/cm2的数据：养护时间*234579121417212856抗压强度y354247535965687374828499试求型回归方程。%建立volum.m文件function yhat=volum(beta,*);yhat=beta(1)+beta(2)*log(*);%输入*=2 3 4 5 7 9 12 14 17 21 28 56;y=35 42 47 53 59 65 68 73 76 82 86 99;beta0=5 1;beta,r,J=nlinfit(*,y,volum,bet
5、a0);beta结果：beta = 21.0058 19.5285所得回归模型为：画线：plot(*,y,r-)*=2 3 4 5 7 9 12 14 17 21 28 56;u=log(*);u=ones(12,1) u;y=35 42 47 53 59 65 68 73 76 82 86 99;b,bint,r,rint,stats=regress(y,u);b,bint,stats结果为：b = 21.0058 19.5285bint = 19.4463 22.5653 18.8943 20.1627stats = 1.0e+03 *0.0010 4.7069 0.0000 0.0009
6、做残差图：rcoplot(r,rint)预测及作图：z=b(1)+b(2)*log(*);plot(*,y,k+,*,z,r)1. 设有五个样品，每个只测量了一个指标，分别是1，2，6，8，11，试用最短距离法将它们分类。(样品间采用绝对值距离。) clcclearb=1;2;6;8;11;d=pdist(b,cityblock); D=squareform(d);z=linkage(d); H=dendrogram(z);T=cluster(z,2); 结果：各样品之间的绝对距离为：距离矩阵,样品间的最短距离为：；2.表1是1999 年中国省、自治区的城市规模构造特征的一些数据，试通过聚类分
7、析将这些省、自治区进展分类。(表1见下页省、自治区城市规模万人城市首位度城市指数基尼系数城市规模中位值万人京津冀699.71.43710.93640.780410.88179.461.89821.00060.58711.78111.131.4180.67720.515817.775389.61.91820.85410.576226.32211.341.7881.07980.456919.7052592.30590.34170.507623.48沪923.193.7352.05720.620822.16139.291.87120.88580.453612.67102.781.23330.5326
8、0.379827.375108.51.72910.93250.468711.12129.23.24541.19350.451917.08173.351.00180.42960.450321.215151.541.49270.67750.473813.94434.467.13282.44130.528219.19139.292.35010.8360.48914.25336.543.54071.38630.40222.195*96.121.22880.63820.514.3445.432.19150.86480.41368.73川渝365.011.68011.14860.57218.6151466
9、.63332.37850.535912.25136.222.82791.29180.598410.4711.794.15141.17980.61187.315244.045.11941.96820.628717.8145.494.75151.93660.580611.6561.368.26950.85980.80987.4247.61.50780.95870.48439.73*128.673.85351.62160.490114.47a=699.70001.43710.93640.780410.8800179.46001.89821.00060.587011.7800111.13001.418
10、00.67720.515817.7750389.60001.91820.85410.576226.3200211.34001.78801.07980.456919.70502592.30590.34170.507623.4800923.19003.73502.05720.620822.1600139.29001.87120.88580.453612.6700102.78001.23330.53260.379827.3750108.50001.72910.93250.468711.1200129.20003.24541.19350.451917.0800173.35001.00180.42960
11、.450321.2150151.54001.49270.67750.473813.9400434.46007.13282.44130.528219.1900139.29002.35010.83600.489014.2500336.54003.54071.38630.402022.195096.12001.22880.63820.500014.340045.43002.19150.86480.41368.7300365.01001.68011.14860.572018.61501466.63332.37850.535912.2500136.22002.82791.29180.598410.470
12、011.79004.15141.17980.61187.3150244.04005.11941.96820.628717.8000145.49004.75151.93660.580611.650061.36008.26950.85980.80987.420047.60001.50780.95870.48439.7300128.67003.85351.62160.490114.4700;d1=pdist(a); %欧氏距离：d1=pdist(a);,%b中每行之间距离z1=linkage(d1) %作谱系聚类图：H= dendrogram(z1)T=cluster(z1,3) % 输出分类结果结
13、果为：1z1 = 8.0000 15.0000 1.6521 20.0000 24.0000 2.0877 18.0000 26.0000 2.4880 11.0000 27.0000 2.7654 21.0000 28.0000 3.9199 29.0000 32.0000 6.9926 3.0000 10.0000 7.1673 13.0000 33.0000 7.3528 31.0000 35.0000 8.6125 2.0000 12.0000 11.2916 9.0000 34.0000 12.7262 17.0000 38.0000 12.8051 25.0000 30.0000
14、15.5084 6.0000 23.0000 16.3291 36.0000 39.0000 18.0388 37.0000 42.0000 22.9979 4.0000 19.0000 25.7717 16.0000 44.0000 28.7559 5.0000 43.0000 32.8508 41.0000 46.0000 32.9368 22.0000 40.0000 33.7288 47.0000 48.0000 36.1367 14.0000 45.0000 45.7490 49.0000 50.0000 77.5676 1.0000 7.0000 223.7891 51.0000 52.0000 265.43562输出分类结果：T = 1 3 3 3 3 3 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3说明，假设分三类，3是一类，2是一类，其它的是一类。
