Is it possible to forecast timeseries using wavelets?

Question

I have a bunch of minute data that have many seasonalities (day, week, month, year). I want to make a long forecast of it using wavelet transforms.

By this moment I've made a fourier transform and got an expression for predicting the daily seasonality. So I've extracted frequencies and amplitudes for sine functions. Seasonal trend (orange) and reconstructed seasonal trend(blue) The blue curve is the sum of 5 sine functions with known amplitudes and frequencies. And if I fit x-values to this equation I get y-values.

Also I'm trying to make the same thing with wavelet transformation (using pywt).

import numpy as np
import pywt
x = np.linspace(0, 1, num=2048)
chirp_signal = np.sin(250 * np.pi * x**2)
data = chirp_signal

fig, ax = plt.subplots(figsize=(40,10))
ax.set_title("Original Signal: ")
ax.plot(data)
plt.show()

waveletname = 'sym5'

fig, axarr = plt.subplots(nrows=5, ncols=2, figsize=(40,10))
for ii in range(5):
    (data, coeff_d) = pywt.dwt(data, waveletname)
    axarr[ii, 0].plot(data, 'r')
    axarr[ii, 1].plot(coeff_d, 'g')
    axarr[ii, 0].set_ylabel("Level {}".format(ii + 1), fontsize=20, rotation=90)
    axarr[ii, 0].set_yticklabels([])
    if ii == 0:
        axarr[ii, 0].set_title("Approximation coefficients", fontsize=20)
        axarr[ii, 1].set_title("Detail coefficients", fontsize=20)
    axarr[ii, 1].set_yticklabels([])

plt.tight_layout()
plt.show()

The example is from this guide.

And the question is: Is it possible (and if "yes" then how) to make a function with received coefficients, fit x-values to it and get corresponding y-values?

Thanks!