Benefits and Drawbacks of the Stock Market Production

The Stock Market is a challenging forum for investment and requires immense brainstorming before one shall put their hard earned money to work. This project aims at processing large volumes of data and running comprehensive regression algorithms on the dataset; that will predict the future value of a stock using the regression model with the highest accuracy. The purpose of this paper is to analyze the shortcomings of the current system and building a model that would mitigate most of them by implementing more efficient algorithms. Using this model, anyone can monitor the preferred stock that they want to invest in; and maximize profit by purchasing volume at the lowest price and liquidating the stock when it’s at its highest.

Introduction

A stock (also known as 'shares' or 'equity') is a type of security that signifies proportionate ownership in the issuing corporation. This entitles the stockholder to that proportion of the corporation's assets and earnings. Stocks are bought and sold predominantly on stock exchanges, though there can be private sales as well, and are the foundation of nearly every portfolio. These transactions have to conform to government regulations which are meant to protect investors from fraudulent practices. Historically, they have outperformed most other investments over the long run. These investments can be purchased from most online stock brokers or at the stock exchange. Stock market prediction is the act of trying to determine the future value of a company stock or other financial instrument traded on an exchange. The successful prediction of a stock's future price could yield significant profit. The efficient-market hypothesis suggests that stock prices reflect all currently available information and any price changes that are not based on newly revealed information thus are inherently unpredictable. Others disagree and those with this viewpoint possess myriad methods and technologies which purportedly allow them to gain future price information.

Stock market price data is generated in huge volume and it changes every second. Stock market is a complex and challenging system where people will either gain money or lose their entire life savings. In this work, an attempt is made to predict the stock market trend. We take the current stock values from the data sets gathered. The data gathered is modelled into various sub parts or data sets which is used to train and test the algorithm. We use regression models in python or R to model the data. We run a comprehensive search algorithm on the data sets and create a summary table based on the output. We plot the values on a chart and apply regression and clustering techniques to find out the increase or decrease in price of that stock. Based on the calculation, we extrapolate the current stock prices to generate a prediction after a given time. Supervised machine learning algorithms are used to build the models. The output will be in graphical form and will change with change in dataset. We expect up to 66% of in-sample accuracy and 35% or out-of-sample accuracy using supervised machine learning algorithms on prediction model. This will enable the user to take better decisions while investing in the stock market. We attempt to minimize erroneous results by implementing Aggregation in order to mitigate bad predictions or waste.

Related Work

Stock market is considered the primary indicator of a country’s economic strength and development. Stock Market prices are volatile in nature and are affected by factors like inflation, economic growth, etc. Prices of a share market depend heavily on demand and supply. High demanded stocks will increase in price whereas heavily sold stocks will decrease in price. Fluctuating stock prices affects the investor’s belief and thus there is a need to predict the future stock value. The objective of this review is to predict the stock market prices in order to make more informed and accurate investment decisions. Recent trends in stock market prediction are surveyed; different types of machine learning classifiers and their respective variants are applied on them. Various approaches and the results of past years are compared based on methodologies, datasets and efficiency and then it is represented in the form of a Graph. The survey describes different theories and conventional approaches to stock market prediction. Along with it, it discusses recent machine learning techniques along with pros and cons of each technique for effectively predicting the future stock prices followed by various researchers.

Limitations Of The Present System

The present system only focuses on Linear Regression, which is not as accurate as other regression models. This is because, while performing Linear Regression, we assume that the dependent and independent variables are linearly correlated. However, that is not the case, every time. Therefore, not only may it be infeasible in certain circumstances, the dataset needs to be tailored to the specific algorithm to get an accurate outcome. The present system is seemingly short-sighted. This is because it focuses on only a single stock value but ignores the various other non-linear parameters that may exist and my affect the accuracy of the model. It does not talk about the relative accuracy of similar prediction models. Therefore, it is safe to say that there may exist a more accurate outcome.

Benefits Of The Proposed System

The project will implement various regression algorithms on the same dataset, finding out the relative accuracies of the different models. The most accurate model will be deployed to predict the stock prices, working on the given dataset. It will also take into consideration those parameters, that are not linearly correlated; therefore, expanding the scope of the system and eliminating the issues of short-sightedness. The range of applicability of the system will also be expanded to such an extent that it can run over various datasets and provide an equally accurate outcome. The efficiency of the system is expected to be second to none. Graphical outputs will provide excellent visualization of the outcome, which will make it easy to understand, even for novice users.

Trade-Offs Of The Proposed System

Certain limitations of the proposed system that are expected to be seen, are as follows: While solving the problem of redundancies, we are avoiding the usage of intangible parameters that may influence the stock market; like Human sentiments, social media influence, reputation of the firm, etc. There may exist models that implement these parameters to get a more accurate outcome. However, these are beyond the scope of the project. Acquiring real-time data from the stock market can be quite cumbersome. Therefore, to avoid unnecessary issues of implementation, it has been excluded of the scope of the project.

Technologies Implemented

Predictive modelling is a process that uses data mining and probability to forecast outcomes. Each model is made up of a number of predictors, which are variables that are likely to influence future results. Once data has been collected for relevant predictors, a statistical model is formulated. The model may employ a simple linear equation, or it may be a complex neural network, mapped out by sophisticated software. As additional data becomes available, the statistical analysis model is validated or revised. Regression is a statistical measurement used in finance, investing and other disciplines that attempts to determine the strength of the relationship between one dependent variable (usually denoted by Y) and a series of other changing variables (known as independent variables). Regression helps investment and financial managers to value assets and understand the relationships between variables, such as commodity prices and the stocks of businesses dealing in those commodities.

• Linear Regression: Y = a + bX + u
• Multiple Linear Regression: Y = a + b1X1 + b2X2 + b3X3 +... + btXt + u
• Multiple Linear Regression with interaction.

Regression takes a group of random variables, thought to be predicting Y, and tries to find a mathematical relationship between them. This relationship is typically in the form of a straight line (linear regression) that best approximates all the individual data points. In multiple regression, the separate variables are differentiated by using numbers with subscripts.

Conclusion

The aim of our research study is to help the stock brokers and investors for investing money in the stock market. The prediction plays a very important role in stock market business, which is very complicated and challenging process due to dynamic nature of the stock market. This project aims at finding the best prediction model among a plethora of those existing today, and implementing the one with the highest empirical and/or real accuracy in order to predict stock prices. The purpose of this paper is to analyze the shortcomings of the current system and building a model that would mitigate most of them by implementing more efficient algorithms. This would indeed make investment in the stock market, a safer bet.