Internet Searches Trend and Forecast

In this report, we will try to analyze academic program trends in the United States by extracting google searches between 07-2018 and 08-2020. Then, we will implement a forecasting procedure with R language, based on an additive model where non-linear trends are fit with yearly, weekly, and daily seasonality.
First, we can check the google searches for “Nursing Programs” searches:

Then, we do the same for “Technology Programs” searches:

Additionally, to display a better comparison chart, we will extract searches for another type of medical school programs “Technician Programs”.

Now, Let’s summarize the searches per each category in order to have a clearer vision of what we’re looking at:

Internet Searches Forecast

Finally, we can see that the results are even clearer when they are shown together according to academic program type. Using the information from the last 2 years, we will attempt to predict internet searches per academic program type for the next year. In order to do this, we implement a forecasting procedure with R language, based on an additive model where non-linear trends are fit with yearly, weekly, and daily seasonality, plus holiday effects:

Quick Conclusions:

According to the forecast, google searches for these programs will slightly decrease over the next year. Some possible causes:
  • People are changing search habits (social networks, other search engines, etc).
  • They are looking for another programs.
  • Data from this unusual year has a negative impact on the forecast algorithms.