Introduction
Theophrastus (a Greek philosopher) once said “time is the most valuable thing a man can spend”. Time is valuable because it is scarce. It is so valuable in fact that there is a whole subsection of mathematics dedicated to it – time-series analysis. With the advent of data analysis applicability to sectors of the economy where time is a very prominent factor, there has evolved to be a subsection of data analysis dedicated to time series analysis. But what actually is time series analysis and what are the sectors that use it day in and day out? That’s exactly what the article “What Is Time Series Data?” by 365 Data Science on medium explains (the article can be found here).
Summary of Article
The article starts out by giving a definition of time series analysis which is the following: analysis that has to deal with a time series (“a sequence of information that attaches a time period to each value”). The only stipulation is that the time series has to have a start point and endpoint. Then the article goes into some other definitions of related topics. They are as follows:
- Frequency – “how often the values of the data set are recorded”
- Patterns – “patterns observed in time series are expected to persist in the future.”
Next, the article gets into applications of time-series-analysis in 2 primary sectors – weather prediction and business. It is apparent why time is important in these two sectors because in weather prediction you are looking at the weather for a certain interval in time and in business, especially finance, you are looking at huge amounts of money in the stock market over time. Then the article gets into time dependency. This is the phenomenon where the values of every period are also affected by outside factors but also past values. The article gives an example of how tomorrow’s weather outside will be about the same as today. Finally, the article states its last concept of denoting time series. For this, we use capital letters such as X or Y for time series variables. We use an uppercase “T” to denote the entire time period used in analysis and a lowercase “t” to describe a single period within a time series interval. The article concludes by giving some final words.
My Take
Overall, I really liked this article. I have been seeing this phrase a lot nowadays and this article really helped to clear some things up. It makes sense why there is a subsection of data analysis dedicated to this as time is used as a variable in a lot of studies and analyses. I do wonder if there are any special ways to treat this analysis compared to other analysis sectors because it is time, are there any extra variables that you have to keep in mind when doing your analysis? Or is there any more functionality you are granted because time is continuous and a number, so you can manipulate it by doing operations on it. I do really like working with time as the independent variable because, especially if it is times that you have lived through, it provides a certain level of relatability that is unmatched by any other category of analysis and it is fun to see how differnt things progressed over time and try to find reasons for that. I will definitely be looking more into this topic.
Conclusion
All in all, this article was a really good one. I really liked the concept and how it was explained. It sparked a plethora of questions that I will definitely be looking more into. I highly recommend you read it (the article can be found here)