In life, it helps to learn what not to waste one’s time on. So is the case in data science. For example, it’s useful to know the long term sales trend of a newly launched product.

In life, it helps to learn what not to waste one’s time on. So is the case in data science.

For example, it’s useful to know the long term sales trend of a newly launched product. It’s not very useful to know the sales trend from one second to the next.

The variation in sales at this level of granularity may well be, indistinguishable from white noise.

And yet, there happens to be a statistical model to do just that — modeling of white noise. It is known as the white noise model, and it goes like this for time series data:

The additive white noise model

In this model, we are saying that the value observed at time step *i* is the summation of the current level of the time series at time step *i* and a random component *N*i_ around the current level.

If the extent of random variation is proportional to the current level, then we have the following multiplicative version of the same model:

The multiplicative white noise model

If the current level *L*i_ is constant for all *i*, i.e. *L*i = L_ for all *i*, then the noise will be seen to fluctuate around a fixed level.

It’s easy to generate a white noise data set. Here’s how to do it in Excel:

time-series-analysis programming data-science noise python data analysis

🔵 Intellipaat Data Science with Python course: https://intellipaat.com/python-for-data-science-training/In this Data Science With Python Training video, you...

🔥Intellipaat Python for Data Science Course: https://intellipaat.com/python-for-data-science-training/In this python for data science video you will learn e...

Master Applied Data Science with Python and get noticed by the top Hiring Companies with IgmGuru's Data Science with Python Certification Program. Enroll Now

What is a Time Series? Depending on the frequency of observations, a time series may typically be hourly, daily, weekly, monthly, quarterly and annual. Sometimes, you might have seconds and minute-wise time series as well, like, number of clicks and user visits every minute etc.

I knew I wanted to do two things in the process of writing my bachelor’s thesis: improve my programming skills and work with time-series data prediction.