{"title":"Determination of Rolling Returns from the Performance Chart of Mutual Funds","authors":"Jia-Hao Chew, C. Ng","doi":"10.1109/CSPA52141.2021.9377307","DOIUrl":null,"url":null,"abstract":"Mutual funds are a popular form of investment and savings. The performance chart and returns table are available from the factsheet of the fund, which can be used to evaluate the performance of the fund. However, the rolling returns, which is not given, may be a better indicator of the fund's long-term performance consistency. A semi-automated software application was developed to determine the rolling returns of a mutual fund from its published performance chart. Because the published chart is in rasterized image format, image processing was applied to extract the data. Some manual input is required from the user, including the initial and final dates, values and points, and loading an image of the performance chart. The image is converted into hue-saturation-value format and cropped. Thresholding and morphological operations are then performed to obtain a binary image. The graph is detected by finding the contours and the x-y data points extracted. The data, together with the information input by the user, are used to calculate the rolling returns. From our evaluation, the average error for 5-year rolling returns was an absolute difference of 3.63 between the actual and calculated percentage values.","PeriodicalId":194655,"journal":{"name":"2021 IEEE 17th International Colloquium on Signal Processing & Its Applications (CSPA)","volume":"60 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-03-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2021 IEEE 17th International Colloquium on Signal Processing & Its Applications (CSPA)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CSPA52141.2021.9377307","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Mutual funds are a popular form of investment and savings. The performance chart and returns table are available from the factsheet of the fund, which can be used to evaluate the performance of the fund. However, the rolling returns, which is not given, may be a better indicator of the fund's long-term performance consistency. A semi-automated software application was developed to determine the rolling returns of a mutual fund from its published performance chart. Because the published chart is in rasterized image format, image processing was applied to extract the data. Some manual input is required from the user, including the initial and final dates, values and points, and loading an image of the performance chart. The image is converted into hue-saturation-value format and cropped. Thresholding and morphological operations are then performed to obtain a binary image. The graph is detected by finding the contours and the x-y data points extracted. The data, together with the information input by the user, are used to calculate the rolling returns. From our evaluation, the average error for 5-year rolling returns was an absolute difference of 3.63 between the actual and calculated percentage values.