K-MEANS AND K-MEDOID ALGORITHM APPLICATION IN CLUSTERING STOCK DATA IN INDONESIA

Abstract

Stock or share is a sign of a person's or a business party's capital participation in a company. In the context of stock, the meaning of investment itself is an activity of investing funds or capital in a company. The investment is expected to generate profits and not vice versa, therefore a good consideration is needed in the selection of the company's shares to be purchased. Parameters that can be used as the basis for consideration are the level of volatility, liquidity, and also market capitalization. Volatility is an indicator of the rate of change in stock prices that implies the risk of how the stock market price moves. The greater the level of volatility, the greater the risk of the stock market. Liquidity is an indicator of how easy the shares are to be sell and purchased without affecting the asset price. Meanwhile, stocks with high market capitalization will be difficult to play with. Therefore we need to know the characteristics of stocks before investing our fund. The results of data clustering using K-Means and K-Medoid help us to understand the characteristics of stocks. With the effectiveness of the clustering algorithm, we can see the characteristics of stocks, from those with a high risk, high level of liquidity, and also could calculate the average market capitalization in the clustering result for each cluster. The results of this analysis can be the basis for good stock selection, based on the
stock historical data.