CODEWITHSUNDEEP
python
RedZoro
RedZoro

Reputation: 13

Q-Q plot in python eror in the theorical quantile axe

I need to plot a QQ graph with the following information:

spcs2k = np.array([[ 49, 524,  16,  87, 157,  58,   4,  41, 110,  90,   2,  41, 136,
        495, 249,  40,  48,   3,  72, 294,  49,  28, 163,  61,  89,   2,
        168, 286,  23,  67,  19,  11,  63,   4, 246, 130,   2, 378, 176,
        251,  78, 138,  97,  34,  33, 183,  12, 209,  82,  87,   9,  33,
         19,  77,  54,  28,  59,  88, 202,  12,  53,  86, 146,  26, 112,
        176,  35,  94, 180,  93,   8,  32,  26,   5, 145,  13,   5, 138,
        205,  42,  17, 134,  19,  54, 133, 134,  10, 173,   3,  59, 223,
        109, 175, 266, 314,  68, 283,  71,  77, 147,  32,  70, 131, 112,
         32,  29,  19,  28,  85,  25,  57,  16, 130, 157,  13, 167,  29,
          2, 442,  10, 150, 185,  95,  57,  63, 150,  41,  22,  72,  59,
          2,   8,   5, 156,  51, 161, 243, 152, 289,  93,  34, 140,  74,
         34,  37,   9, 121, 138,  94,  67,  65, 202,  67,  13, 240, 209,
          2, 296,   6,  61,   2, 134, 196]])

import statsmodels.api as sm
import scipy.stats as stats
from matplotlib import pyplot as plt

fig = sm.qqplot(spcsk2,stats.expon,line="45")
plt.show()

but i get this: enter image description here

and the idea is get a similar graph like this:

enter image description here

thanks for supporting me

Upvotes: 0

Views: 1222

Answers (2)

xen20
xen20

Reputation: 63

Whoever has this problem should pass fit=True to the sm.qqplot method. This will auto-determine parameters such as loc, scale, and distargs. See docs here: https://www.statsmodels.org/dev/generated/statsmodels.graphics.gofplots.qqplot.html .

Upvotes: 2

totnan
totnan

Reputation: 65

The code works fine, it does what it should. QQ plot show if the data that you pass to it is normally distributed or not. In your case this means that the values are not even vaguely normally distributed in spcs2k.

If you run this code, you can see what it looks like on a dataset that came from normal distribution.

data = np.random.normal(0,1, 1000)
fig = sm.qqplot(data, line='45')
plt.show()

Upvotes: 1

Related Questions