Version 1
: Received: 5 September 2022 / Approved: 6 September 2022 / Online: 6 September 2022 (04:21:52 CEST)
How to cite:
E., E.G.; E., A.C.; T, A.S.; R., V.; R., G.K. On the Impacts of Hard Data Patterns on Bayesian Maximum Entropy Performance: Simulation-based Analysis. Preprints2022, 2022090082. https://doi.org/10.20944/preprints202209.0082.v1
E., E.G.; E., A.C.; T, A.S.; R., V.; R., G.K. On the Impacts of Hard Data Patterns on Bayesian Maximum Entropy Performance: Simulation-based Analysis. Preprints 2022, 2022090082. https://doi.org/10.20944/preprints202209.0082.v1
E., E.G.; E., A.C.; T, A.S.; R., V.; R., G.K. On the Impacts of Hard Data Patterns on Bayesian Maximum Entropy Performance: Simulation-based Analysis. Preprints2022, 2022090082. https://doi.org/10.20944/preprints202209.0082.v1
APA Style
E., E.G., E., A.C., T, A.S., R., V., & R., G.K. (2022). On the Impacts of Hard Data Patterns on Bayesian Maximum Entropy Performance: Simulation-based Analysis. Preprints. https://doi.org/10.20944/preprints202209.0082.v1
Chicago/Turabian Style
E., E.G., Vihotogbé R. and Glèlè Kakaï R.. 2022 "On the Impacts of Hard Data Patterns on Bayesian Maximum Entropy Performance: Simulation-based Analysis" Preprints. https://doi.org/10.20944/preprints202209.0082.v1
Abstract
Bayesian Maximum Entropy (BME) is increasingly used in predicting and mapping spatio-temporal data. However, studies that have fully evaluated its robustness empirically are rare. Therefore, this research examined empirically the effect of skewness, sample size and spatial dependency level using simulated data. We considered symmetric data, data positively skewed by 1, 3, 6 and 9, data with weak, moderate, and strong spatial dependency levels, and sample sizes from 100 to 500 at the interval length of 50. The results showed that the variation of sample sizes and spatial dependency levels do not affect the Mean Square Error (MSE) and bias of BME prediction. However, skewness affects the MSE of prediction but does not affect the bias. This result indicates that BME is robust to sample size and is unbiased. Despite the significant difference due to skewness, a graphical plot showed values of MSE close to zero, suggesting that BME can be considered robust to skewness.
Keywords
samples size; spatial dependency; skewness; Bayesian Maximum Entropy
Subject
Environmental and Earth Sciences, Environmental Science
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.