You are on page 1of 8

More on polynomials.

>> x=-4.0:0.25:6.0;
>> p=[2, 3, 4, -5, -6, 3]; % 2x^5 + 3x^4 + 4x^3 -5x^2 -6x +3
>> v=polyval(p,x);
>> roots(p)
ans =
-0.8954 + 1.6642i
-0.8954 - 1.6642i
-1.0654
0.9342
0.4220
plot(x,v,'LineWidth',3,'color','r')
>>
4

2.5

x 10

1.5

0.5

-0.5
-4

-3

-2

-1

Regression curve and polynomial fit:


>> n=2;
>> q=polyfit(x,v,n)

% quadratic curve fitting

q=
1.0e+003 *
0.3369 0.4999 -1.6000
>> n=1;
>> q=polyfit(x,v,n)

% linear line fitting

q=
1.0e+003 *
1.1738 1.0111
>> R=corrcoef(x,v);
>> R(1,2)

% obtains correlation coefficient R

ans =
0.7214
>>
>> clear
>> sales = round(100*rand(30,4))
sales =
81
91
13
91
63
10
28

71
3
28
5
10
82
69

75
26
51
70
89
96
55

8
5
53
78
93
13
57

55
96
96
16
97
96
49
80
14
42
92
79
96
66
4
85
93
68
76
74
39
66
17

32
95
3
44
38
77
80
19
49
45
65
71
75
28
68
66
16
12
50
96
34
59
22

14
15
26
84
25
81
24
93
35
20
25
62
47
35
83
59
55
92
29
76
75
38
57

47
1
34
16
79
31
53
17
60
26
65
69
75
45
8
23
91
15
83
54
100
8
44

>> sorted_sales=sort(sales)
sorted_sales =
4
10
13
14
16
17
28

3
3
5
10
12
16
19

14
15
20
24
25
25
26

1
5
8
8
8
13
15

39
42
49
55
63
66
66
68
74
76
79
80
81
85
91
91
92
93
96
96
96
96
97

22
28
28
32
34
38
44
45
49
50
59
65
66
68
69
71
71
75
77
80
82
95
96

26
29
35
35
38
47
51
55
55
57
59
62
70
75
75
76
81
83
84
89
92
93
96

16
17
23
26
31
34
44
45
47
53
53
54
57
60
65
69
75
78
79
83
91
93
100

>> days=1:30;
>> subplot(2,2,1);
>> plot(days, sales(:,1))
>> subplot(2,2,2);
>> plot(days, sales(:,2))
>> subplot(2,2,3);
>> plot(days, sales(:,3))
>> subplot(2,2,4);
>> plot(days, sales(:,4))
>>

100

100

80

80

60

60

40

40

20

20

10

20

30

100

100

80

80

60

60

40

40

20

20

10

20

30

10

20

30

10

20

30

Also,
>> R=corrcoef(sales)
R=
1.0000
-0.0636
-0.2455
0.1253

-0.0636
1.0000
-0.0070
-0.1968

-0.2455
-0.0070
1.0000
-0.0851

0.1253
-0.1968
-0.0851
1.0000

>> R=corrcoef(sorted_sales)
R=
1.0000 0.9634 0.9512 0.9464

0.9634 1.0000 0.9891 0.9880


0.9512 0.9891 1.0000 0.9936
0.9464 0.9880 0.9936 1.0000
>>
The sorted sales profiles of city 3 and city 4 are highly correlated.
>> % Some statistical functions.
>> x=rand(1,80); % A row of 80 random number between 0 and
1.0
>> x=100*rand(1,80); % A row of 80 random number between 0
and 100.0
>> avg = mean(x)
avg =
49.5715
>> sigma_x = std(x)
sigma_x =
27.2316
>> median_x = median(x)
median_x =
47.9766
>> variance= var(x)
variance =

741.5610
>> sqrt(variance)
ans =
27.2316
>>
>> clear
>> clear
>> x=linspace(-1.0,0.2,50);
>> y=humps(x);
>> plot(x,y)
>> x=linspace(-1.0,2.5,50);
>> y=humps(x);
>> plot(x,y)
100

80

60

40

20

-20
-1

-0.5

0.5

1.5

2.5

>> % Integrate function from -1.0 to 2.5


>> area=trapz(x,y); % Use trapezeum rule to evaluate the integral
>> area
area =

23.7579
>>

You might also like