docs/Summary.md

In saradindusengupta/vedicDateTime-GSoC-2022: Demo R package for VedicDateTime (part A) for GSoC, 2022

Summary

For the article titled "Natural Time-Series Analysis and Vedic Hindu Calendar System"

Time is the most crucial component in time sequence based forecasting where seasonality and trend changes based on the time component of the data. Since all of the business-related data points are recorded based on the solar calendar, i.e. Gregorian calendar, it misses one of the major influencing factors on earth, which is the moon or lunar time. Any calendar is either solar, lunar or lunisolar, that is the time component is referenced either against how long it takes for the earth to revolve around the sun, how long it takes the moon to revolve around the earth or both. Since the Gregorian calendar is a solar-based calendar many natural phenomena such as lunar rhythms which are responsible for influencing animal and bird activities such as mating period, migration, including lunar periods into the time component would provide better results in those scenarios especially.

Apart from this, since the length of the months in the Gregorian calendar is not equal, situations, where total days are counted, make divergent calculations because the length of months is different from the previous one and the next one. For example, businesses using full load scheduling for work are at odds due to this difference in the number of days between months. Since the Vedic calendar or Hindu calendar divides the year into 12 months of equal lengths and adds one month after 30 months to adjust for the day count it would be easier for regular operations in many businesses to have an equal-length month. Moreover, since in Vedic or Hindu calendar manages days based on lunar days, i.e., taken by the Moon to increase its longitude by 120 over the earth's longitude, the length of the days (or ‘thithis’ as it is called) is not equal but rather varies between 20 to 27 across the year.

Since no conclusive studies have been done to analyse the effect of both the scenarios in actual use-cases to understand the accuracy or synchronous patterns emerging with time-series segmentation, the practical use-cases cannot be considered. One point of insight that can be concluded just from analysing daily temperature data over 1 year period is that the correlation coefficients between each month's periods are higher in the Vedic or Hindu calendar than it is in the Gregorian calendar. This observation gives a preliminary stand to analyse the use-cases for lunisolar style calendars, for this scenario Vedic or Hindu calendar.

saradindusengupta/vedicDateTime-GSoC-2022 documentation built on April 16, 2022, 12:53 a.m.