SAS菜鸟求助！

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

最近在阅读文献中，有这么一段我不知道他是如何记录原始数据并编辑SAS分析程序的，请高手指点。原文：Aphid Response to BYDV-Infected Versus Noninfected Untransformed Varieties ‘Lambert’ and ‘Caldwell’.The response of aphids to BYDV-infected versus noninfected plants of Lambert and Caldwell was tested in separate arenas. There were 10 replicates for each comparison per variety. The arena was made from a 150-mm-diameter polystyrene petri dish. The bottom of the petri dish was lined with 150-mm-dimeter Þlter paper (Whatman), which was in contact with wheat leaves. The leaves, attached to the test plants, were positioned horizontally in the dish, equally distant from the center opposite to one another. Three leaves from each of two treatments (infected or noninfected) were used to produce a dual-choice test. Aphids could touch the leaves with their tylets or tarsi. Aphid locations on or near the leaves (1 cm on either side of leaves) were recorded every 20 min for a 2-h period. The total number of aphids tested for the entire bioassaywas 800. Each test was analyzed as a separate experiment comparing the number of aphids on each treatment and using a generalized linear model, assuming a binomial distribution with a logit link function (PROC GENMOD; SAS nstitute 1990). Aphids not located on either treatment were excluded from the analysis.谢谢！

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏0 回帖

关键词：horizontally distribution transformed Generalized Generalize logit link function

I am not sure if my guess is correct.
As you mentioned GENMOD with logit link and binomial distribution, I guess the dependent variable is on which leaf the aphid stays. I name the dependent variable as BYDV-Infected, by assigning BYDV_Infected=1 if the aphid is on the leave of BYDV-Infected, and BYDV_Infected=0 if the aphid is on the leaf that was uninfected.

800 aphids are used in this experiment, and every 20 minutes a record is made to write down if the aphid stays on the infected leaf or on the uninfected leaf. This implies the existence of repeated measurements. Hence I suppose each aphid has a unique identification number which is aphid_id.
Let’s suppose the independent variables are age of the aphid, sex of the aphid, size of the aphid and species of the aphid, and the name of the data is bioassay.
The SAS code in this situation for a GEE1 model is:

proc genmod data=bioassay descending;
class species sex;
model  BYDV_Infected = age sex size species / link=logit dist=bin type3;
output out=out p=p;
repeated subject= aphid_id /type=cs;
run;

Perhaps you can attach the full article, and let's see if I can explain more for help.

3# eugeniayao