Sample data created with python

Sample data

Linear data

n=20

a = np.arange(n).reshape(4, -1); a  #5-column matrix

array([[ 0,  1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15, 16,
        17, 18, 19, 20, 21, 22, 23, 24],
       [25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41,
        42, 43, 44, 45, 46, 47, 48, 49],
       [50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66,
        67, 68, 69, 70, 71, 72, 73, 74],
       [75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91,
        92, 93, 94, 95, 96, 97, 98, 99]])

df = pd.DataFrame(a, columns=list('abcde')); df

Random data

r = np.random.randn(4, 5); r

array([[-0.37840346, -0.84591793,  0.50590263,  0.0544243 ,  0.59361247],
       [-0.2726931 , -1.74415635,  0.0199559 , -0.20695113, -1.19559455],
       [-0.59799566, -0.26810224, -0.18738038,  1.05843686,  0.72317579],
       [ 1.23389386,  1.91293041, -1.33322818,  0.78255026,  2.04737357]])

df = pd.DataFrame(r, columns=list('abcde')); df

df.plot()

<matplotlib.axes._subplots.AxesSubplot at 0x17699af2a58>

df = pd.DataFrame(np.random.randn(n,n))

plt.contourf(df, cmap='jet')

<matplotlib.contour.QuadContourSet at 0x1769a1a12b0>

Contour line display

plt.pcolor(df, cmap='jet')

<matplotlib.collections.PolyCollection at 0x1769b1e2208>

Color map display

sin wave

n=100
x = np.linspace(0, 2*np.pi, n)

s = pd.Series(np.sin(x), index=x)
s.plot()

<matplotlib.axes._subplots.AxesSubplot at 0x1769e695780>

sin wave

snoise = s + 0.1 * np.random.randn(n)
sdf = pd.DataFrame({'sin wave':s, 'noise wave': snoise})
sdf.plot(color=('r', 'b'))

<matplotlib.axes._subplots.AxesSubplot at 0x1769e8586d8>

I put noise on it

normal distribution

from  scipy import stats as ss

median = x[int(n/2)]  #Median of x
g = pd.Series(ss.norm.pdf(x, loc=median), x)
g.plot()

<matplotlib.axes._subplots.AxesSubplot at 0x1769ffba128>

gnoise = g + 0.01 * np.random.randn(n)
df = pd.DataFrame({'gauss wave':g, 'noise wave': gnoise})
df.plot(color=('r', 'b'))

<matplotlib.axes._subplots.AxesSubplot at 0x1769e970828>

log function

median = x[int(n/2)]  #Median of x
x1 = x + 10e-3
l = pd.Series(np.log(x1), x1)
l.plot()

<matplotlib.axes._subplots.AxesSubplot at 0x1769ffba5f8>

lnoise = l + 0.1 * np.random.randn(n)
df = pd.DataFrame({'log wave':l, 'noise wave': lnoise})
df.plot(color=('r', 'b'))

<matplotlib.axes._subplots.AxesSubplot at 0x176a00ec358>

Random walk

n = 1000
se = pd.Series(np.random.randint(-1, 2, n)).cumsum()
se.plot()

<matplotlib.axes._subplots.AxesSubplot at 0x284f3c62c18>

Random walk is drawn by randomly generating n of (-1, 0, 1) with np.random.randint (-1, 2, n) and accumulating them with cumsum ().

sma100 = se.rolling(100).mean()
ema100 = se.ewm(span=100).mean()

df = pd.DataFrame({'Chart': se,  'SMA100': sma100, 'EMA100': ema100})
df.plot(style = ['--','-','-'])

<matplotlib.axes._subplots.AxesSubplot at 0x284f3cadcc0>

A Simple Moving Average and an Exponential Moving Average were drawn at the same time. It is generally said that EMA is easier to reflect the latest movements and follow trends than SMA.

It has nothing to do with the content of the article, but if you write it in jupyter notebook and drop it in md format, it's really easy because you only have to attach it to qiita.