1 / 24

LATIN SQUARE

LATIN SQUARE. Erlina Ambarwati. LATIN SQUARE. Used if there are two source of variation among experimental unit Two directional blocking: column and row Treatment=row=column Impractical (disadvantage). Perbedaan: Antar tandan & antar sisir. . A. . B. . . C. D. Sumber keragaman:

march
Download Presentation

LATIN SQUARE

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. LATIN SQUARE Erlina Ambarwati

  2. LATIN SQUARE • Used if there are two source of variation among experimental unit • Two directional blocking: column and row Treatment=row=column Impractical (disadvantage) Erlina Ambarwati

  3. Perbedaan: Antar tandan & antar sisir  A  B   C D • Sumber keragaman: • Antar Tandan • Antar Sisir  E Erlina Ambarwati

  4. Perbedaan kesuburan dan pencahayaan Beda Pencahayaan B E D A K E S U B U R A N Erlina Ambarwati

  5. Tata letak dasar LS B D C A D C B A B A C D D B C A A B C D Erlina Ambarwati

  6. B D C A D C B A B A C D D B C A Pengacakan Latin Square /Kolom Kolom diacak: 3, 2, 4, 1

  7. C D A D C A C A D D A C Pengacakan LS (baris) B B C B B C B B C B B C A=1 B=2 Perlakuan diacak: 1, 3, 2, 4 C=3 D=4 Erlina Ambarwati

  8. Latin Sqaure Row IA B C D Row IIC D A B Row III D C B A Row IV B A D C Column 1 2 3 4 Erlina Ambarwati

  9. Linear Model for LS i = 1, 2, 3, ….., t j = 1,2, 3, ……, t k = 1, 2, 3, ….., t Erlina Ambarwati

  10. Berbagai Nilai duga dengan model Erlina Ambarwati

  11. Erlina Ambarwati

  12. Xijk - =ζi + αj + βk +εikj, ζi , αj, βk& εijk independen   (Xijk - )2=  (ζi +αj + βk + εijk)2   (Xij - )2=  ζi2 + ∑∑ αi2 + βj2+ εijk2 JK total JK perlakuan JK Baris JK Kolom JK sesatan   (Xijk - )2=   Xijk2 + t22 – 2t22 =   Xijk2 – t22 =   Xijk2 – t2 (  Xij)2/(t)2 =   Xijk2 – (  Xij)2/t2

  13. JK perlakuan JK plk =t  i2 Erlina Ambarwati

  14. JK BARIS JK baris =t  αj2 Erlina Ambarwati

  15. JK KOLOM JK baris =t  βk2 Erlina Ambarwati

  16. Erlina Ambarwati

  17. Analysis of Variance Table for a LATIN SQUARE Statistical Decision: when Ho is true, MSTr/MSE ~ F;(t-1);,(t-1)(t-2) if VRF reject Erlina Ambarwati

  18. Example 1: Erlina Ambarwati

  19. ANOVA Conclusion: There are significant diferrent between treatment effect Erlina Ambarwati

  20. Missing Data pada LS R= jumlah pengaruh baris C= jumlah pengaruh kolom T= jumlah pengaruh treatment G= grand total (jumlah total) t = jumlah treatment atau jumlah kolom atau jumlah baris Erlina Ambarwati

  21. Konsekuensi adanya missing data untuk uji lanjut dg LSD a. Jika tidak ada missing data b. Mengandung satu missing data c. Mengandung dua missing data t1 dan t2 : angka ulangan efektif ErlinaAmbarwati

  22. Cara mencari t1 dan t2 Angka ulangan efektif Erlina Ambarwati

  23. Contoh Untuk A = 0 + 2/3 + 1 + 1 + 2/3 = 10/3 Untuk B = 2/3 + 2/3 + 1 + 1 + 0 = 10/3 Erlina Ambarwati

  24. SOAL LATIHAN Penelitian menggunakan Rancangan Latin Square dengan tiga perlakuan (G,H dan I) yang mempunyai bentuk dasar, randomisasi dan hasil sebagai berikut: Bentuk dasar a.Tunjukkanrandomisasibaris, randomisasikolomdan randomisasiperlakuannya. b. Hitung data hilangdengan meminimalkanJksesatan. *) data hilang ErlinaAmbarwati

More Related