Abstract
This study aims to examine the application of the High-Low method, Scatterplot, and Least Squares in cost analysis, particularly in separating fixed and variable cost behavior. The research uses a library research approach by collecting and reviewing literature from textbooks, journals, and official publications related to cost estimation and cost behavior modeling. The findings indicate that the High-Low method is simple because it uses only the highest and lowest activity levels, but its accuracy is limited due to reliance on extreme data points. The Scatterplot method provides a visual representation of the relationship between cost and activity, helping identify patterns and potential outliers; however, its results may vary due to subjective judgment in determining the cost line. Meanwhile, the Least Squares method is considered the most accurate because it utilizes all available data to produce the best-fit regression line, although it requires more complex calculations and statistical understanding. Overall, the selection of the most suitable method depends on the objective of the analysis, data availability, and the level of accuracy required