Which is the best statistical method to compare observed and predicted results.
My data is on laboratory soil column studies for transport of 5 compounds in subsoil for different soil conditions, leak rates. Same was simulated using ANSYS. Which is the best way to statistically compare my data. Thank you.
Correlation coefficient between two datasets can be used to analyse accuracy of predicted data. Also as stephen sir said a plot between observed and predicted data may be made and if both are in good agreement it will fit along the 45 degree line. Artificial neural networks and genetic algorithm can be used in case of complex datasets to develop an equation which best fits the data. For simple datas correlation can be obtained by doing a regression analysis in excel
My statistical knowledge is very poor. I have set of data's for 5 compounds transporting at different rates in soil for different set of experimental conditions observed from laboratory studies. Similar conditions were simulated in Ansys Fluent. I would like to present my results in a meaningful manner by correlating chemical properties of 5 compounds and physical characteristics of soil for different seasons. Hence I am looking forward for statistical methods to achieve the same.
Thank you,
Perhaps of some interest for your query:
https://www.researchgate.net/post/how_to_explain_the_difference_between_simulation_and_experiment_result
- Badges
- Science topic
- Similar topics
- Social Psychology
- Attitude