8382

7763

كارشناسي ارشد

آمار اقتصادي و اجتماعي

اصفهان: دانشگاه صنعتي اصفهان، دانشكده علوم رياضي

1392

نه،108ص.: مصور،جدول،نمودار

ص.ع.به فارسي و انگليسي

سعيد پولادساز

سروش عليمرادي

15/10/92

هوشنگ طالبي، غلامرضا اميدي

رياضي

ID7763

به فارسي و انگليسي: قابل رويت در نسخه ديجيتالي

Optimality and e ciency of circular neighbour balanced block designs for correlated observation Mansoure Motallebi Fesharaki m motallebi@math iut ac ir 16 September 2013 Department of Mathematical Sciences Isfahan University of Technology Isfahan 84156 83111 Iran Supervisor Dr Saeid Pooladsaz spooladsaz@cc iut ac ir Advisor Dr Soroush Alimoradi salimora@cc iut ac ir 2010 MSC 62Kxx 62K05 62K10 Keywords Autoregressive process E ciency Correlated errors Information matrix Interferencemodel Neighbour designs Universal optimalty AbstractThe assumptions in the classical block model are that the response on a plot to a particular treatmentdoes not a ect the response on the neighbouring plots and the treatments associated with plots in ablock is constant In many experiments especially in agriculture the response on a given plot may bea ected by treatments on neighbouring plots as well as by the treatment applied to that plot Therefore the estimates of treatment di erences may deviate because of interferencing of neighbouring units Neighbour balanced block designs are used for modeling and controlling interference e ects betweenneighbouring plots witch all the designs are assumed such as linear blocks with neighbour e ects onlyin the direction of the blocks say left neighbour and right neighbour e ects As the e ect of havingno treatment di ers from the neighbour e ect of any treatment so we consider only designs withborder plots On the other hand all the designs are circular that is the treatment on a border plotis the same as the treatment on the inner plot at the other end of the block Assume two distinct models Y 1n Td Ld B e Y 1n Td Ld Rd B e

سعيد پولادساز

سروش عليمرادي

هوشنگ طالبي، غلامرضا اميدي