Naslov
Approaches to Data Transformations and Their Impact on the Skewness Statistic For Seriously Skewed Distributions: Selected Cryptocurrencies' Data Explored

Autori
Dumičić, Ksenija ; Cunjak Mataković, Ivana

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
Proceedings of the 16th International Symposium on Operational Research in Slovenia SOR'21 / Drobne, Samo ; Zadnik Stirn, Lidija ; Kljajić Borštnar, Mirjana ; Povh, Janez ; Žerovnik, Janez - Ljubljana : Slovensko društvo informatika, 2021, 441-446

ISBN
978-961-6165-57-0

Skup
16th International Symposium on Operations Research in Slovenia (SOR '21)

Mjesto i datum
Bled, Slovenija, 22.09.2021. - 24.09.2021

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
Cryptocurrencies, data transformation, hypothesis testing, normal distribution, seriously skewed distribution, significance level

Sažetak
The normality of distribution presents the underlying assumption of many statistical methods. In some cases, the observed data do not always follow a symmetric bell-shaped distribution or approximately normal distribution. Therefore, data transformation is a commonly used approach to deal with skewed data distributions. This paper analyses the three most often used data transformations: square root transformation, log transformation, and inverse transformation, and their impact on seriously skewed distributions. The proposed data transformations applied to several cryptocurrencies resulted in different statistically significant outcomes, which might be useful for quantitative analysis of data and prognostic modelling.

Izvorni jezik
Engleski

Znanstvena područja
Ekonomija

POVEZANOST RADA