Taylor Diagram Software
We have used Taylor Diagram Software in these papers:

Rainfed wheat (Triticum aestivum L.) yield prediction using economical, meteorological, and drought indicators through pooled panel data and statistical downscaling
Prediction of effective climate change indicators using statistical downscaling approach and impact assessment on pearl millet (Pennisetum glaucum L.) yield through Genetic Algorithm in Punjab, Pakistan
Formulas
Tutorial
Download
License
Github
You can access sample input files for download here: DropBox

What is Taylor Diagram software tool?
When assessing the quality of multiple models compared to observation data and conducting various statistical tests, the process can be complex, especially when simultaneously dealing with several models, scenarios, and parameters. In such cases, using a Taylor diagram software tool proves to be the most efficient approach to achieving accurate results in a shorter time.

Agrimetsoft recognized the need for a Taylor diagram software tool and explored existing options, including programming languages like NCL (Ncar Command Language / Taylor diagram NCL), MATLAB (MATLAB Taylor diagram), Python (python Taylor diagram), and various R packages (R Taylor diagram) that enable the creation of Taylor diagrams. However, for those unfamiliar with coding in these software environments, utilizing these options can become a tedious challenge.

To address this issue, Agrimetsoft developed its own Taylor diagram software, specifically designed to streamline the process for scholars and scientists. This software provides a user-friendly interface, making it easy and efficient to generate Taylor diagrams. With the Agrimetsoft Taylor diagram software, users can effortlessly create Taylor diagrams with multiple models and input variables, saving valuable time. The software also offers the flexibility to save the diagrams in various file formats, ensuring high-quality outputs.

Before ordering the Taylor diagram software, it is recommended to consult us and use the demo version(You can input your data and not draw the chart) and watch the accompanying YouTube video, which provides step-by-step guidance on running the software. By utilizing this tool, researchers can simplify the process of depicting Taylor diagrams, ultimately enhancing their analysis and facilitating their work in scientific research.

How to Draw Taylor Diagram | Windows Software
What is Taylor Diagram?
Taylor diagrams are a valuable tool for evaluating multiple models and comparing their performance across multiple aspects. They provide a graphical summary of how well a set of models matches observations by considering the standard deviation of model values' time series and the correlation between the model values and observations. Taylor (2001) introduced these diagrams as a comprehensive means of assessing model performance, incorporating information such as root mean square error (RMSE) and correlation coefficients.

Taylor diagrams allow for a visual comparison of various variables from one or more test data sets against reference data sets. Typically, the test data sets consist of model experiments(Model/Predicted data), while the reference data set serves as a control or includes observational data. The plotted values on the diagram are often derived from climatological monthly, seasonal, or annual means. To account for differing numerical values across variables, normalization is applied using the reference variables. The ratio of normalized variances reflects the relative amplitude of variations between the models and observations.

In a Taylor diagram, various statistical indicators for each model are combined within a single quadrant. The x-axis represents the RMSE (Root Mean Square Error) or normalized RMSE, while the y-axis represents standard deviation values or their normalized counterparts. The internal semi-circles on the diagram correspond to both RMSE and standard deviation values. Additionally, the arc axis represents the correlation coefficient, and the correlation coefficient values continue as diagonal lines into the graph.

These indicators in Taylor Diagram collectively assess the performance of each model in comparison to the observations. Ideally, a strong linear relationship between the model and observations is indicated by a correlation coefficient close to one. Greater accuracy is reflected by smaller RMSE values, and the standard deviation of the model results should closely align with that of the observations.

RMSE serves as a measure of the discrepancies between predicted values and observed values, providing an assessment of accuracy. On the other hand, the correlation coefficient quantifies the strength of the linear relationship or pattern similarity between two variables. It is commonly used to gauge the extent of correlation between variables. Another advantage of Taylor diagrams is their ability to plot different parameters together by normalizing the indicators(RMSE, SD), enabling a comprehensive evaluation across multiple dimensions. This second type of Taylor diagram is recommended for analysis

How to Draw Taylor Diagram with Negative Correlation Values
Taylor Diagram Software
How can we compute Taylor Diagram?
In essence, the Taylor diagram provides a statistical representation of the relationship between two fields: a "test" field, typically representing a model simulation, and a "reference" field, typically representing observed data. Taylor (2005) discovered that each point in the two-dimensional space of the Taylor diagram can simultaneously depict three key statistics: the centered RMS difference, the correlation coefficient, and the standard deviation. This is achieved through a simple formula that relates these statistics to the Taylor diagram.

The construction of the Taylor diagram is based on the similarity between the equation mentioned earlier and the Law of Cosines. In the equation, R represents the correlation coefficient between the test and reference fields, E'2 denotes the centered RMS difference between the fields, and σ f^{2} and σ r^{2} represent the variances of the test and reference fields, respectively. By considering the correlation coefficient as the cosine of the azimuthal angle, the Taylor diagram takes shape, leveraging the resemblance to the Law of Cosines.

There are several minor variations of the Taylor diagram that have proven to be useful for different purposes. For more in-depth information, I recommend referring to Taylor's original paper from 2001. It provides detailed insights into these variations and their specific applications.

References:
IPCC, 2001: Climate Change 2001: The Scientific Basis, Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change [Houghton, J.T., Y. Ding, D.J. Griggs, M. Noguer, P.J. van der Linden, X. Dai, K. Maskell, and C.A. Johnson (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 881 pp.
Elvidge, S., Angling, M.J., et al., 2014. On the use of modified Taylor diagrams to compare ionospheric assimilation models, 978-1-4673-5225-3/14/$31.00, 2014 IEEE.
Taylor KE. 2001. Summarizing multiple aspects of model performance in a single diagram. Journal of Geophysical Research 106: 7183-7192.
Taylor Diagram Primer, 2005, Karl E. Taylor. January 2005