The Analysis of Stock Market Trends Using The K-Nearest Neighbor (KNN) Algorithm

The Analysis of Stock Market Trends Using The K-Nearest Neighbor (KNN) Algorithm

For many years, stock forecasting has been a challenging subject for experts in finance and

statistics. In general, there are two methods for predicting the stock market. One of them is fundamental

analysis, which is reliant on a business’s operating procedures and foundational information. The

performance of the supervised machine learning algorithm KNN (K-Nearest Neighbor) is evaluated by the

author in this study. The K-Nearest Neighbor machine learning technique is used in this study to anticipate

using prices with daily and minute frequency, the three separate markets’ stock prices for high and small

capitalizations. Forecasting trends for the stock market are thought to be a significant and effective

activity. As a result of intelligent investing decisions, stock prices will result in favourable returns. Due to

the static and noisy data, investors find it challenging to make predictions about the stock market. As a

result, investors who seek to increase the value of their investment have a significant challenge when trying

to anticipate the stock exchange. Making analysis about the stock market involves the use of mathematical

techniques and educational technology. In this article, past and suggested procedures are presented,

including computation techniques, machine learning calculations, and execution elements. This research is

therefore helping to uncover datasets and machine learning strategies for stock market Forecasting. The

most often used techniques for producing precise stock market forecasts are ANN and NN. The most modern

method of forecasting the stock market has many shortcomings despite the extensive work that went into it.

In this perspective, stock showcase estimation is seen as a coordinates handle, therefore specific factors for

doing so should be given more consideration. The economies of developed countries are assessed in terms of

their power economy. Stock markets are currently regarded as a prestigious trading industry since, in many

circumstances, they provide simple gains The most modern method of forecasting the stock market. The

stock market/exchange, with its extensive and dynamic data sources, is seen to be an ideal setting for data

mining and business researchers. To help investors, managers, decision-makers, and users make wise and

informed investment decisions, we forecasted stock prices for a list of the companies that are listed on the

stock market/exchange using the k-nearest neighbour method and a non-linear regression technique. The

results demonstrate the robustness and low error ratio of the KNN approach, leading to sensible and

reasonable results. Furthermore, the forecast results were near to and practically parallel to actual stock

prices, based on the actual stock price data. Because the stock market has such a huge influence on a

country’s economy, it’s intriguing to observe how stock market prediction may be employed and whether or

not the expected outcomes are accurate. Using the closing prices of shares listed on the Stockholm stock

market OMX, this research will compare the prediction approaches, the kNN 9K Nearest Neighbour)

algorithm, and the MA (moving average) formula. The paper covers these theoretical principles allowing

the reader to obtain a good familiarisation of the backdrop of stock markets and the formulae used. With

the help of appropriate charts and tables, the data is distributed properly. Finally, a commentary highlights

the ramifications of the findings, as well as the conclusion that the K Nearest Neighbor algorithm provided

more accurate data than the moving average approach.