Predicting Prices for My New Business
I want to predict prices for my new business. I am curious, and I am a beginner in data. I am trying to do my best. I decided to take a problem that is close to my environment.
I always enjoy looking at wonderful graphs that analyze data. Before this, I studied some cases of stock prices in university.
Let’s look at this example graph.
Beatiful, right?
In this blog post. I just want to share some important steps from my findings with you.
Let’s start.
Data and Datasets
Data is like gold. It is very useful information. After collecting many records, you will have your dataset. Let’s read Wikipedia for some important facts, like code. Amazing!!. You can find amzazing dataset in Kaggle or Open Data.
What Is a Parameter?
A dataset parameter is a customizable field that you can add to a dataset worksheet.
We need define Search Parameters
Search parameters are important for your research question. Consider:
- The frequency of data collection(dates).
- The geographic region(coordinates).
- The time period timesets (one moment in time vs. changes over time).
Searching Samples for our goal
Here are some datasets I previously looked at:
- Tesla stock market data (Date, Open, High, Low, Close, Volume).
- Uber dataset (Date, Open, High, Low, Close, Adj Close, Volume).
- And many more about stock.
What Else Do I Need to Know?
The Process of Cleaning
Since I am a beginner, I use simple cleaning steps.
- In most cases, I replace NaN or null values with the mean.
- If the dataset is very large, sometimes I just remove the missing rows.
Linear Regression
Now, some math comes in.
The simplest form of the regression equation with one dependent and one independent variable is:
$$ y = mx + c $$
- Our target (y) is the “Close” price.
- The other variables are the input (x).
The link to the resource is at the end of this article.
Let’s See the Graph
Graphs are the best way to understand results. I want everybody to see what is happening.
One useful chart is: Plot Actual vs Predicted Prices Over Time.
Conclusion
I want to answer real-life questions using data. Step by step, I will learn and improve.