概率論及sas作業(yè)

1、概率論及SAS作業(yè)P1421在校園中有3盞誘蛾燈，連續(xù)5天誘蛾記錄如下：地點(diǎn)觀測(cè)記錄112463038172112124522313416118試用SAS先做Levene的F檢驗(yàn)，再作對(duì)數(shù)轉(zhuǎn)換后的方差分析。解：（1）檢驗(yàn)方差是否同性質(zhì)，用LeveneF檢驗(yàn)：H0各方差具有相同性質(zhì)sas代碼：datacap_fly;doa=1to3;doi=1to5;inputx;output;end;end;cards;124630381711212452213416118;procanova;classa;modelx=a;meansa/hovtest;run;程序輸出結(jié)果：LevenesTestforHo

2、mogeneityofxVarianceANOVAofSquaredDeviationsfromGroupMeansSumofMeanSourceDFSquaresSquareFValuePrFa255635.727817.94.470.0353Error1274596.96216.4/0.0353Fa20.04230.02110.200.8207Error121.26320.1053SAS至線The油口邛PrDcedLre2J:44Monda：Ceptridtrit在riablE：ySun.SourceDFSquarssLeanSqutrsFlalePrFNodEl22.5023915C1,

3、2544E753.57O.C608Er-cr124.209250JE0.35077C34Co-rectedTotal146.7-21415-SquareCjef-VarootMSEyNear.于289口?1.82362J.5322592.71E8WSourceDFAniwaS8LeanSqutrsFMusPrF方差來(lái)源平方和自由度2均方和K1,25僻值3用顯著性因素A2.5021.253.57N誤差4.21120.35總和6.7114,可見(jiàn)轉(zhuǎn)換后方差同質(zhì)（0.80270.05）,滿足方差同質(zhì)性假設(shè)，方差分析的結(jié)果是接受原假設(shè)，認(rèn)為各地點(diǎn)誘蛾數(shù)量不存在顯著差異（顯著性水平0.05）P1471.對(duì)

4、3個(gè)玉米品種用3種方法進(jìn)行管理，得到9個(gè)小區(qū)的產(chǎn)量（單位：kg）如下:品所-一_方法_B1B2B3A113043A312945A3152547作方差分析，估計(jì)各總體的未知參數(shù).和j及.若有必要，求出兩兩均值差的雙側(cè)0.95置信區(qū)間。解：H01:各i.都相等，H02:各.j都相等作雙因素方差分析，并對(duì)均值差作雙側(cè)區(qū)間估計(jì)。Sas代碼：datacorn;doa=1to3;dob=1to3;inputx;output;end;end;cards;113043312945152547;procanova;classab;modelx=ab;meansab;meansab/lsdcldiff;DDn;S

5、ourceDF怕mSquareFValuePrF36J0J0001046JOJOOO3.000000523.0000001.030.423013.030.0176SAS至婉-he-4Dracecure08:4i-iDependenlvariable：乂SunofSourceDFSquaresffetnSquareFValuePrFMode141I32J0J0002E3.0000007.030.口Error4IGOJOJOOOiO.OOOOOOCorrectedTotal81292JOJOOOR-SquareCoeffVarRootl?SExMeai0.87616120,623556.32455

6、530.66667SAS更統(tǒng)Leve1ofTheANOVAProcedure八8LNMeanStdDev1328.100000016.03347692335.10000008.7177979Leve1of329.100000016.3707055NMeanStdDevb1319.100000010.58300522328.10000002.64575133345.10000002.0000000Compsirisonssignificantatthe0.05IeveIareindicatedby出*小.35笈ConfidenceLimitsDifferenceaBetweenCompariso

7、nMeans2-32-13-23-11-21-36.0007.000-6.0001.000-7.000-1.000888833333333I_H_UI_Lu212120.33821.3388.33815.3387.33813.338ComparisonDifferenceBetweenMeans35笈ConfidenceLimits17.0102.66231.338出帆26.01011.66240.338m-17.010-31.338-2.662m-9.010-5.33823.338-26.010-40.338-11.662m-9.010-23.3385.338方差來(lái)源平方和自由度均方和F值顯

8、著性因素A862431.08N因素B1046252313.08*誤差160440總和12928,各品種間不存在顯著性差異，各管理方法對(duì)產(chǎn)量有顯著性差異。均值估計(jì)結(jié)果品種管理方法產(chǎn)量均值A(chǔ)1A2A3B1B2B330.67283529192845/管理方法B1與B2間均值差的雙側(cè)0.95置信區(qū)間為1-23.338,5.338，管理方法B1與B3間均值差的雙側(cè)0.95置信區(qū)間為1-40.338,11.662，管理方法B2與B3間均值差的雙側(cè)0.95置信區(qū)間為【-31.338,-2.662】P153盆栽小麥?zhǔn)㎞肥施P肥試驗(yàn)得到單株產(chǎn)量（單位：KG）N一一一一一-_P_P0P1N09,12,910,13

9、,10N113,15,1414,20,17做方差分析，估計(jì)總體未知參數(shù)i.和.j及，若有必要，求出兩兩總體均值差雙側(cè)0.95置信區(qū)間。解：H01：施不同量N肥無(wú)顯著差異；H02：施不同量P肥無(wú)顯著差異；SAS代碼：datawheat;don=1to2;dop=1to2;doi=1to3;inputx;output;end;end;end;cards;9129101310131514142017;procanova;classnp;modelx=npn*p;meansnp;meansnp/lsdcldiff;run;程序輸出結(jié)果：SASStSE13:03TuesThe昌NOV島Procedure

10、SourceDFSumofSquaresMeanSquareFValuePrFNode1390.000000030.00000007.500.0104Error832.00000004.0000000CorrectedTota111122.0000000R-8quireCoeffVarRootMSExMean0.73770515.384S22.00000013.00000SourceDFAnovaSSMeanSquareFValuePrFn175.0000000075.0000000018.750.0025P112.0000000012.000000003.000.1215n：+：p13.00

11、0000003.000000000.750.4117SAS羔統(tǒng)TheANOVAProcedureLevsIornNMeanStdDev1610.50000001.643167672615.50000002.53843582LevsIorPNMeanStdDev1612.00000002.529822132614.00000003.94963353Comp-arisonssignificantatthe0-051eve1areindicsl+edb：y。索米CoiiipirisonDifferenceBetreenMeans35常ConfidenceLimits5.000-5.0002.3377

12、.663-7.663-2.337方差來(lái)源平方和自由度均方和F值顯著性N肥7517518.75*P肥121123.0NN、P肥交互作用3130.75N誤差3284總和12211可見(jiàn)，施N肥對(duì)產(chǎn)量有極顯著差異，施P肥和施N、P肥交互作用對(duì)產(chǎn)量無(wú)顯著性影響。均值估計(jì)施N肥施P肥總均值氮肥1氮肥2磷肥1磷肥21310.515.51214/施氮肥1與施氮肥2對(duì)產(chǎn)量均值差的雙側(cè)0.95置信區(qū)間為【-7.663,-2.337】P158研究某品種玉米施N肥對(duì)蛋白質(zhì)含量的影響，安排施肥與不施肥兩個(gè)處理，每處理又有3個(gè)不同小區(qū)取容量為2的樣本進(jìn)行化學(xué)分析：處理施氮肥不施氮肥小區(qū)數(shù)據(jù)1234563,99,1110,

13、128,62,46,4做方差分析，估計(jì)各總體未知參數(shù)i.和ij和。若有必要，求兩總體0.95雙側(cè)置信區(qū)間。解：H0：施氮肥與不施氮肥對(duì)玉米蛋白質(zhì)含量無(wú)顯著差異；HO1：施氮肥或不施氮肥時(shí)各小區(qū)玉米蛋白質(zhì)含量無(wú)顯著差異；Sas代碼：datacorn;don=1to2;dogroup=1to3;doi=1to2;inputx;output;end;end;end;cards;399111012862464;procanova;classngroup;modelx=ngroup(n);meansngroup;meansngroup(n)/lsdcldiff;run;程序輸出結(jié)果：SourceDFSu

14、mofSquaresMeanSquareFVaiuePrFMode1592.00000001S.4COOOOO3.940.0625Error620.00000004.6666667CorrectedTota111120.0000000R-SquareCoeffVarRootUSExMean0.76666730.S60672.16C2477.000000SourceDFftnovaSSMeanSquareFVaiuePrFn140.0000000048.0000000010.290.0184group(n)444.000000001I.OOCOOOOO2.360.1666LeveIotgroupLeve1otMMeanStdDev1126.00000004.2426406921210.00000001.4142135631211.00000001.414213561227.00000001.414213562223.00000001.414213563225.00000001.41421356Comparisonssignificantatthe0.051evelareindicatedby出*.differencenBetween95宣CorfidenceComparisonMean