Python Calculation of Basic Stock Metrics


Any stock can be described using descriptive statistics.


An introduction

A few metrics will be computed in Python in this blog post that provide a brief overview of a specific stock. In researching stocks, a number of factors may influence stock performance; these specific indicators can help you capture a stock's initial movement and compare it to others.

Defining metrics - an overview

Averaging is the sum of all prices divided by the total number of prices available in a period of time.

The simple rate of return.  Basically, you subtract the starting value from the current value and divide it by the starting value. Multiplying a final value by 100 will give you a percentage return.

Daily Returns.  Changes in price from one day to the next, expressed as a percentage.

Standard Deviation.  A measure of the degree to which the values within a set are spread out. Standard deviations are higher when there is a greater range of possible outcomes. In principle, the standard deviation measures how much the values deviate from the mean; the lower the standard deviation, the more similar measurements cluster together. 

Minimum Value.  Your set's lowest value. A price, volume, or other metric could be used here.

Maximum Value. In your set, this value was the highest.

Simple Moving Average.  It's calculated using the average of the closing prices of n recent periods, where n is how many periods we're considering. Using the moving average, you can determine the current trend of a stock by measuring the volatility of the market. Long-term trends can be shifted by changes in short-term moving averages (for example, those of five and twenty days).

Correlation.  A linear relationship between two variables is measured by the strength and direction (positive or negative) of the relationship. In most cases, correlation is calculated using Pearson's r. Correlations that are positive represent a positive linear relationship, while correlations that are negative represent a negative linear relationship. Linear correlation is not present if the value is 0. Linear correlations are stronger when the value is close to 1 or -1. 

Beta.  In relation to a benchmark index, the standard deviation of a stock's historical price movement. Stocks are ranked according to their deviation from the index beta, which is 1.0. Stocks with betas below 1 are less volatile than indices, while stocks with betas above 1 are more volatile.

Technical Examples + Data Download

The data for SPY, the SPDR S&P 500 ETF, will be obtained through yfinance. In addition to accessing financial data from Yahoo Finance, yfinance is an open source library.

Using pip install, let's install yfinance package first.

Yfinance library can be imported after the package is installed.


Now let's download SPY data between January 1, 2018 and September 16, 2022, and check the first five rows.


Average Price Calculation

On the 'Close' column, we will run the mean() function


Simple Rate of Return Calculation

In order to calculate a simple return for the period, we need to use the following formula:

Three weeks later, you sold one share of SPY for USD315 after you bought it at USD300. Three weeks would result in a 5% return.

Our data will now be used to calculate the SPY return.


Returns Calculated Daily

The Daily Return will be calculated by adding a new column 'Daily Return' and calculating the following:


Since pct_change() does not return percentage values, the formula is multiplied by 100. Due to the fact that there are no rows before the first, the first row is also NaN.

The Standard Deviation Calculator

To calculate the standard deviation, we will use .std().

Is this number significant?

Taking into account the normal distribution of the data points. Normal distributions demonstrate that values close to the mean (the average price) occur more frequently than values further from the mean.

The standard deviation from the mean is one standard deviation 68% of the time and two standard deviations 95% of the time in a normal distribution.

There are a range of closing prices between USD 393.27 and USD 446.53 within the first standard deviation (26.63).


Calculating minimum and maximum values

The .min() and .max() functions will be used.


Simple Moving Average Calculation

SMA involves adding up the most recent data points and dividing them by the time period. The Series.rolling() method is used to compute SMA. Using the 'window' argument, we can specify any number of periods. Then plot it on a graph for a 5 and 20-day window.


Plotly library will be used for the graph.



Correlation Calculation

In order to calculate correlation, at least two different stock datasets must be used. An ETF that tracks Treasury bonds over 20 years, TLT, will be compared to SPY.


Correlation can be calculated using the .corr() function:


A correlation of 0.8 indicates a strong positive relationship. It is particularly interesting because historically TLT and SPY have been negatively correlated or have been weakly positively correlated.

Beta Calculation

ETFs tracking 500 companies are called index ETFs. Our objective is to calculate the beta of JPMorgan Chase & Co. (JPM) compared with SPY. Obtain the standard deviations of the returns of the two instruments by downloading JPM data and computing the correlation between them. JPM's beta can be calculated by plugging the values into the following equation:


Since we have daily returns for SPY, let's calculate those for JPM. To avoid getting a percentage, let's not multiply it by 100.


Now let's see how the returns correlate.



Next, we will look at the standard deviations of both instruments' returns.


It's finally time for JPM's beta release.


JPM stock is theoretically less volatile than SPY by 12% (1-0.88).

Thank you for taking the time to read this tutorial. I hope you enjoyed it. I appreciate your time.

Comments

Post a Comment

Popular posts from this blog

Does Autocorrect Make Life Better?

Image
  Systemic failures in machine learning: a cautionary tale. Data science can reduce friction and inconvenience in our everyday lives by applying to many products and businesses. Photo by   Gertrūda Valasevičiūtė   on   Unsplash Every device and service we use is embedded with crafted machine learning models. We become ever more free to focus on what matters in life as they tirelessly remove all irritations and burdens from our lives. What are the chances of this becoming a reality? Taking stock of the many ways machine learning fails us in everyday life is crucial if we ever hope to realize the potential of these technologies. Among the many forms of psychopathy manifest in chatbots are racist image classifiers, sexist recruitment tools, and racist image classifiers. We should instead consider one of the most common forms of machine learning failure, one which affects minorities just as much as majorities: autocorrection. Digital assistance can be as simple as autoco...
Read more

Neural Networks from Scratch For Beginner and Also For Experts

Image
NN-from-Scratch Neural Networks from Scratch ¶ Chia-Hung Yuan & DataLab Fall 2020 In this tutorial, you will learn the fundamentals of how you can build neural networks without the help of the deep learning frameworks, and instead by using NumPy. Creating complex neural networks with different architectures in Python should be a standard practice for any Machine Learning Engineer and Data Scientist. But a genuine understanding of how a neural network works is equally as valuable. This is what we aim to expand on in this article, the very fundamentals on how we can build neural networks, without the help of the frameworks that make it easy for us. Model architecture ¶ We are building a basic deep neural network with 3 layers in total: 1 input layer, 1 hidden layers and 1 output layer. All layers will be fully connected. We implement ReLU and sigmoid activation functions. SGD and Momentum optimizer are available. Let'...
Read more

G2Net Basic audio data augmentation inference

Image
01bbf1f2e5254c9998dc3e291afc2043 Copied From Medium Click here to Read More On Medium! Subscribe to ONEPAGECODE Newsletter. COLAB = False if COLAB == True : from google.colab import drive drive.mount( '/content/drive' ) % cd '/content/drive/MyDrive/Colab Notebooks/kaggle/G2Net2022/code' ! pip3 install timm - q WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv import numpy as np import pandas as pd import matplotlib.pyplot as plt import time import h5py import timm import torch import torch.nn as nn import torchaudio import torchvision.transforms as TF from tqdm.auto import tqdm from sklearn.model_selection import KFold from sklearn.metrics import roc_auc_score from timm.scheduler import CosineLRS...
Read more