Anda di halaman 1dari 37

Ig.

Aris Dwiatmoko

DISTRIBUSI NORMAL BIVARIAT


APA ITU DISTRIBUSI NORMAL BIVARIAT?

 Ingat distribusi Normal Univariat

X = banyaknya batang korek per bungkus

1  x 
2

1   
2  
f ( x)  e
 2
STEP BY STEP DISTRIBUSI NORMAL BIVARIAT
APLIKASI DISTRIBUSI NORMAL BIVARIAT

Hubungan antara tinggi anak laki-laki dan tinggi orang tuanya


Sumber:
 Galton, Francis (1889). Natural Inheritance, MacMillan,
London.
 Galton, F.; J.D. Hamilton Dickson (1886). “Family Likeness
in Stature”, Proceedings of the Royal Society of London,
Vol. 40, pp.42-73.
Tinggi anak laki-laki dan tinggi orang tua rata-rata
75

74

73

72

71
Tinggi anak laki-laki

70

69
Adult Child

68

67

66

65

64

63

62

61

60
63 64 65 66 67 68 69 70 71 72 73
Tinggi rata-rata orang tua
Mid-Parent
250

200

150
FREKUENSI
Frequency

100

50

0
63,5 64,5 65,5 66,5 67,5 68,5 69,5 70,5 71,5 72,5

Tinggi orang tua rata-rata


Tinggi anak laki-laki
180

160

140

120
Frequency

100

80

60

40

20

0
61,2 62,2 63,2 64,2 65,2 66,2 67,2 68,2 69,2 70,2 71,2 72,2 73,2 74,2
Height
DISTRIBUSI BERSAMA TINGGI ANAK LAKI-LAKI DAN
ORANG TUA

1   1    y1  1 2 2   y1  1  y2   2   y2   2 2  
f ( y1 , y2 )  exp          y1 , y2  
2   1  
2
1
2
2  2
 
  2 1  
2
   2
1  
1 2  22  

Bivariate Normal Density


where :
1  E (Y1 )  12  V (Y1 )
 2  E (Y2 )  22  V (Y2 )
E Y1  1 Y2   2 

 1 2
0.15-0.2
0.1-0.15
0.05-0.1
0.2
0-0.05

0.15

0.1

0.05

0
-3 -2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 2.5 3
x1

120 121 0.4


Joint Density Function

0.035-0.04
0.03-0.035
0.025-0.03
62.96
0.02-0.025
64.206
0.04 0.015-0.02
0.035 65.452 0.01-0.015

0.03 66.698 0.005-0.01

67.944 0-0.005
0.025 y1
0.02 69.19
0.015 70.436
0.01 71.682
0.005
72.928
0

 Distribusi Normal Bivariat merupakan generalisasi dari


distribusi Normal Univariat. Keduanya dibangun dari konsep
distribusi frekuensi relatif 1 variabel dan 2 variabel
BERBAGAI BENTUK DISTRIBUSI NORMAL BIVARIAT
http://personal.kenyon.edu/hartlaub/MellonProject/Bivariate2.html

Batang-batang histogram 3 dimensi


DISTRIBUSI NORMAL BIVARIAT DALAM ANALISIS KORELASI

 Korelasi antara dua variabel menyangkut


distribusi probabilitas bersama antara kedua
variabel.
 Oleh karenanya, asumsi Normalitas Bivariat
menjadi syarat bagi validitas analisis korelasi
KORELASI PEARSON
Sheskin, David. Handbook of parametric and nonparametric statistical procedures.
(ISBN 1-158488-440-1

Pearson product-moment correlation coefficient is one of


a number of measures of correlation or association
discussed in this book. Measures of correlation are not
inferential statistical tests, but are, instead, descriptive
statistical measures which represent the degree of
relationship between two or more variables. Upon
computing a measure of correlation, it is common practice
to employ one or more inferential statistical tests in order
to evaluate one or more hypotheses concerning the
correlation coefficient. The hypothesis stated below is the
most commonly evaluated hypothesis for the Pearson
product-moment correlation coefficient.
ASUMSI-ASUMSI
The Pearson product-moment correlation coefficient is based on the following
assumptions:
a) The sample of n subjects for which the value r is computed is randomly
selected fiom the population it represents;
b) The level of measurement upon which each of the variables is based is
interval
or ratio data.
c) The two variables have a bivariate normal distribution. The assumption of
bivariate normality states that each of the variables and the linear combination of
the two variables are normally distributed. With respect to the latter, if every
possible pair of data points is plotted on a threedimensional plane, the resulting
surface (which will look like a mountain with a rounded peak) will be a three-
dimensional normal distribution (i.e., a three-dimensional structure in which any
cross-section is a standard normal distribution). Another characteristic of a
bivariate normal distribution is that for any given value of the X variable, the
scores on the Y variable will be normally distributed, and for any given value ofthe
Y variable, the scores on the Xvariable will be normally distributed.
d. Related to the bivariate normality assumption is the assumption of
homoscedasticity. Homoscedasticity exists in a set of data if the
relationship between the X and Y variables is of equal strength across
the whole range of both variables. Tabachnick and Fidell(1989, 1996)
note that when the assumption of bivariate normality is met, the two
variables will be homoscedastic.
e) Another assumption of the Pearson product-moment correlation
coefficient, referred to as nonautoregression, is discussed in many
books on business and economics. This latter assumption, which is
discussed within the framework of a special case of correlation
referred to as autocorrelation, is only likely to be violated when pairs
of numbers which are derived fiom a series of n numbers are
correlated with one another
PENGUJIAN HIPOTESIS
Null versus Alternative Hypotheses

Upon computing the Pearson product-moment correlation


coefficient, it is common practice to determine whether the
obtained absolute value of the correlation coefficient is
large enough to allow a researcher to conclude that the
underlying population correlation coefficient between the
two variables is some value other than zero.
Null hypothesis HO :  = 0
(In the underlying population the sample represents, the
correlation between the scores of subjects on Variable X
and Variable Y equals 0.)
LANGKAH-LANGKAH PENGUJIAN HIPOTESIS
Tetapkan Hipotesis Nol
dan Alternatif Hitung statistik uji
berdasarkan sampel

UJI bivariat normal

Korelasi
T
Spearman Memenuhi?

Y
Statistik uji

BUAT KESIMPULAN

Tentukan Kriteria
Penolakan Ho
PENGUJIAN ASUMSI NORMAL BIVARIAT
LANGKAH-LANGKAH PENGUJIAN NORMALITAS BIVARIAT

1. Hitung jarak mahalanobis

2. Urutkan jarak tersebut dari terkecil ke terbesar

Di mana m banyaknya sampel


3. Andaikan adalah nilai tabel chi-square
yang memenuhi
4. Buat plot grafik
5. Bila plot grafik tersebut membentuk garis
linear maka distribusi variabel adalah multiva-
riat (perhatikan bila n=2 berarti bivariat)
6. Untuk meyakinkan, grafik tersebut dapat diuji
linearitasnya dengan mencari koefisien
determinasi dari
CONTOH HASIL PENGUJIAN

m=24, m=2
UJILAH NORMALITAS BIVARIAT DATA BERIKUT
BB LP
72 99
76 90
70 94
71 82
75 88
76 97
55 75
47 73
52 82
53 81
74 96
85 96
92 104
LANGKAH-LANGKAH DENGAN SPSS

1. Masukkan data (x,y) yang akan diuji


normalitas bivariatnya.
2. Tentukan jarak mahalanobis:
1. Pilih Analyze --> Regression --> Linear
2. Pindahkan dependen dan independen variabel ke
kotak yang sesuai
3. Klik “Save”, and klik “Mahalanobis”
4. Klik OK.
OUTPUT JARAK MAHALANOBIS (MAH_1)
BB LPMAH_1
72 99 0.01064
76 90 0.01064
70 94 0.26596
71 82 0.52128
75 88 0.01064
76 97 0.68085
55 75 2.08511
47 73 2.72340
52 82 0.52128
53 81 0.68085
74 96 0.52128
85 96 0.52128
92 104 2.39.62
 3. Urutkan jarak mahalanobis
 Blok data ketiga variabel – klik data – sort cases
 Sort by – masukkan variabel mahalanobis distance
 Sort order – klik ascending
 OK
URUTKAN JARAK MAHALANOBIS
BB LP MAH_1
76 90 0.01064
75 88 0.01064
70 94 0.26596
71 82 0.52128
52 82 0.52128
74 96 0.52128
85 96 0.52128
76 97 0.68085
53 81 0.68085
72 99 106.383
55 75 208.511
92 104 239.362
47 73 272.340
4. Hitung dengan n = 13 pengamatan,

• Buat variabel i dan diisikan bilangan 1 sampai dengan 13


• Klik transform-compute variable
• Target variable–isikan v
• Numeric expression (i-0.5)/13
• OK
INI PROSESNYA ...............................
PERHITUNGAN CHI SQUARE

 Transform-compute variable
 Target variable isikan chisquare

 Numeric expression – pilih all (pada function


group) – pilih Pdf.Chisq (pada function and
special variables)- klik -> masukkan pada
numeric expression- tanda (?,?) diganti (2,v)
 OK
INILAH SELENGKAPNYA
HASILNYA,.....................................
 5. Buat plot dan cari koefisien
determinasinya
 Analyze-regression-curve estimation

 Masukkan dependen variabelnya chisquare

 Masukkan independen variabelnya


mahalanobis distance
Lihat OUTPUT SPSS dan Interpretasikan
TERIMA KASIH

Anda mungkin juga menyukai