您好,欢迎访问三七文档
当前位置:首页 > 商业/管理/HR > 企业财务 > 第3章――第2节 回归分析《计量地理学》(华东师大,徐建华)
§3.2回归分析方法一元线性回归模型多元线性回归模型非线性回归模型的建立方法一、一元线性回归模型定义:假设有两个地理要素(变量)x和y,x为自变量,y为因变量。则一元线性回归模型的基本结构形式为式中:a和b为待定参数;为各组观测数据的下标;为随机变量。记和分别为参数a与b的拟合值,则(3.2.2)式代表x与y之间相关关系的拟合直线,称为回归直线;是y的估计值,亦称回归值。bxay(3.2.1)xbayˆˆˆ(3.2.2)aˆbˆyˆn,1,2,①参数a与b的最小二乘拟合原则要求yi与的误差ei的平方和达到最小,即②根据取极值的必要条件,有③解上述正规方程组(3.2.4)式,得到参数a与b的拟合值:niiininiiiibxayyyeQ121122min)()ˆ(niiiiniiixbxaybxay110)(0)((3.2.4)iyˆ参数a、b的最小二乘估计(3.2.3)niiniiixxxyxxyyxxLLb121)())((ˆxbyaˆˆ2112111)(1))((1niiniininiiniiiixnxyxnyx(3.2.5)(3.2.6)显著性检验①方法:F检验法。②总的离差平方和:在回归分析中,表示y的n次观测值之间的差异,记为可以证明(3.2.9)式中,Q称为误差平方和,或剩余平方和,而称为回归平方和。niiyyyyLS12)(总niiyyyyLS12)(总xyxxniiniiniiibLLbxxbxbabxayyU21221212)()()ˆ((3.2.9)niniiiiUQyyyy1122)ˆ()ˆ((3.2.8)③统计量F④F越大,模型的效果越佳。统计量F~F(1,n-2)。在显著水平α下,若FFα,则认为回归方程效果在此水平下显著。一般地,当FF0.10(1,n-2)时,则认为方程效果不明显。2nQUF(3.2.10)二、多元回归模型回归模型的建立①多元线性回归模型的结构形式:②回归方程:在(3.2.12)式中,b0为常数,b1,b2,…bk称为偏回归系数。偏回归系数的意义是,当其它自变量都固定时,自变量每变化一个单位而使因变量平均改变的数值。aakaaaxxxyk22110(3.2.11)kkxbxbxbby22110ˆ(3.2.12)③偏回归系数的推导过程:根据最小二乘法原理,的估计值应该使由求极值的必要条件得方程组(3.2.14)式经展开整理后得min)]([)ˆ(122211012nakakaaanaaaxbxbxbbyyyQ),,2,1(0)ˆ(20)ˆ(2110kjxyybQyybQnajaaajnaaa(3.2.13)).,2,1,0(kii)(k,1,2,,0iib(3.2.14)方程组(3.2.15)式称为正规方程组。引入矩阵:nanaakanakkanakaakaanakananaaanakkaanaaaanaananaaanakkaanaaanaananaanakkanaaayxbxbxxbxxbxyxbxxbxbxxbxyxbxxbxxbxbxybxbxbxnb11122121101112122122121012111112121121011111212110)(....)()()(.)()()()()()()()()()()((3.2.15)knnnkkxxxxxxxxxxxxX2132313222121k211111.11knnnkkkknkkknnTxxxxxxxxxxxxxxxxxxxxxxxxXXA213231322212121113212232221113121111111111nakanakaanakaanakanakaanaanaaanaanakaanaaananaanakanaanaaxxxxxxxxxxxxxxxxxxxxxn12121111212212112111211211111211nyyyY21nbbbbb210则正规方程组(3.2.15)式可以进一步写成矩阵形式BAbnaakanaaanaaanaanknkkknnTyyyxyxyyyyyxxxxxxxxxxxxYXB112111321321223222111312111111求解得:引入记号:YXYXBAbTT11)(),,2,1,())((1kjixxxxLLnajjiiajiij),,2,1())((1kiyyxxLnaaiiaiy(3.2.16)正规方程组也可以写成:kkkykkkkkykkykkxbxbxbybLbLbLbLLbLbLbLLbLbLbL2211022112222212111212111)51.2.3(回归模型的显著性检验①回归平方和U与剩余平方和Q:②回归平方和:③剩余平方和为:④F统计量为:计算出来F之后,可以查F分布表对模型进行显著性检验。k21x,,x,xQULSyy总nanaiyiLbyyU112)ˆ(nayyaaULyyQ12)ˆ()1/(/knQkUF三、非线性回归模型非线性关系线性化的几种情况:①对于指数曲线,令,可以将其转化为直线形式:,其中,;②对于对数曲线,令,,可以将其转化为直线形式:;③对于幂函数曲线,令,,可以将其转化为直线形式:其中,;bxdeyxbayxbaylnxbaybdxyxbayyylnxxdalnyyxxlnyylnxxlndaln④对于双曲线,令,转化为直线形式:;⑤对于S型曲线,可转化为直线形式:;⑥对于幂乘积:,只要令,就可以将其转化为线性形式:其中,;xbay1xbayxxexyybeay,1,1令xbaykkxxdxy2121kkxxxy22110xxyy1,1,ln,,ln,ln,ln2211kkxxxxxxyydln0⑦对于对数函数和只要令,就可以将其化为线性形式:例:下表给出了某地区林地景观斑块面积(Area)与周长(Perimeter)的数据。下面我们建立林地景观斑块面积A与周长P之间的非线性回归模型。kkxxxylnlnln22110kkxxxy22110kkxxxxxxyyln,,ln,ln,2211序号面积A周长P序号面积A周长P110447.370625.39242232844.3004282.043215974.730612.286434054.660289.307330976.770775.7124430833.840895.98049442.902530.202451823.355205.131510858.9201906.1034626270.300968.060621532.9101297.9624713573.9601045.07276891.680417.0584865590.0802250.43583695.195243.90749157270.4002407.54992260.180197.239502086.426266.54110334.33299.729513109.070261.8181111749.080558.921522038.617320.396122372.105199.667533432.137253.335138390.633592.893541600.391230.030146003.719459.467553867.586419.40615527620.2006545.291561946.184198.66116179686.2002960.4755777.30556.9021714196.460597.993587977.719715.7521822809.1801103.0705919271.8201011.1271971195.9401154.118608263.480680.710203064.242245.049614697.1301234.1142469416.7008226.0091624519.867326.3171225738.953498.6566313157.6601172.916238359.465415.151646617.270609.801246205.016414.790654064.137437.3552560619.0201549.871665645.820432.3552614517.740791.943676993.355503.7842731020.1001700.965684304.281267.9512826447.1601246.977696336.383347.136297985.926918.312702651.414292.235303638.766399.725712656.824298.47331585425.10011474.770721846.988179.8663235220.6401877.476731616.684172.8083310067.820497.394741730.563172.1433427422.5701934.5967511303.970881.0423543071.5501171.4137614019.790638.1763657585.9402275.389779277.172862.0883728254.1301322.7957813684.750712.78738497261.0009581.298791949.164228.4033924255.030994.906804846.016324.481401837.699229.40181521457.4007393.938411608.625225.84282564370.80012212.410解:(1)作变量替换,令:,,将上表中的原始数据进行对数变换,变换后得到的各新变量对应的观测数据如下表所示。AylnPxln序号y=lnAx=LnP序号y=lnAx=LnP19.2541066.4383794212.358138.36218629.6787636.4172438.3076225.667487310.340996.6537824410.336376.79791849.1530196.273258457.5084335.3236559.2927427.5528164610.176196.87529469.9773387.168551479.5159096.95184178.838076.03
本文标题:第3章――第2节 回归分析《计量地理学》(华东师大,徐建华)
链接地址:https://www.777doc.com/doc-4045659 .html