NumXL

NumXL

Developer(s)	Spider Financial Corp
Initial release	April 15, 2009
Stable release	1.63.42643.1 / September 30, 2016 (2016-09-30)
Development status	Active
Written in	C, C++
Operating system	Windows
Type	econometrics software
License	proprietary
Website	www.spiderfinancial.com

Numerical Analysis for Excel (NumXL) is an econometrics and time series analysis add-in for Microsoft Excel. Developed by Spider Financial, NumXL provides a wide variety of statistical and time series analysis techniques, including linear and nonlinear time series modeling, statistical tests and others.

Although NumXL is intended as an analytical add-in for Excel, it extends Excel’s user-interface (UI) and offers many wizards, menus and toolbars to automate the mundane phases of time series analysis. The features include summary statistics, test of hypothesis, correlogram analysis, modeling, calibration, residuals diagnosis, back-testing and forecast.

NumXL users have varied backgrounds in finance, economics, engineering and science. NumXL is used in academic and research institutions and industrial enterprises.

User interface

NumXL comes with an elaborate user-interface of menus, toolbars, and interactive wizards, to improve the general usability of the software. The UI components automate some steps in time series analysis and modeling.

Using the UI components and the wizards, a user can specify the time series of interest, fine-tune the desired analysis options and designate the location on a worksheet for the output. NumXL generates the corresponding analysis blocks (with underlying formulas) in the designated location

Function categories

NumXL functions are organized into eleven (11) categories:

Descriptive statistics

• Autocorrelation function (ACF) and partial autocorrelation function (PACF).
• Cross-correlation functions - XCF and EWXCF
• MD, RMD, MAD
• Statistical functions - Calculate the excess kurtosis of the standardized GED and Student's t-dist.
• Empirical distribution functions: histogram, EDF and KDE

Statistical testing

• Hypothesis test for population mean, for standard deviation, for skewness and for excess kurtosis.
• Normality test using Jarque–Bera test, Shapiro–Wilk test, and Chi-square test methods.
• White-noise test - serial correlation tests (Portmanteau test, Ljung–Box test and modified Q-test).
• Autoregressive conditional heteroskedasticity (ARCH) effect test.
• Augmented Dickey–Fuller test (ADF) unit-root test.
• Multicollinearity test.
• Regression stability test (Chow test).

Transform

• Box-Cox and inverse Box-Cox transform.
• Logit, Probit and complementary log log link functions and its inverse.
• Lag or backshift operator,
• Seasonal difference operator
• Time series integral operator
• Add, subtract, scale, and chronological order reverse functions.
• Remove missing values from a time series.

Smoothing

• Weighted moving average (WMA).
• Exponential smoothing functions: Brown's simple exponential, Holt-Winters double exponential, Brown's linear exponential, and Winters triple exponential smoothing functions.
• Time series trend - linear, polynomial, logarithmic, exponential, and power.

Spectral analysis

• Discrete Fourier transform and inverse.
• Hodrick–Prescott filter.
• Baxter-King (BK) filter.

Date and calendar

• Weekday and workday calculations
• International weekend support
• 11 international holidays support.

ARMA modeling

• Autoregressive–moving-average model (ARMA)
• Autoregressive integrated moving average (ARIMA) model
• Seasonal ARIMA (SARIMA)
• AirLine Model
• X-12-ARIMA, Seasonal adjustment (X11)
• Autoregressive moving average exogenous (ARMAX) model
• Seasonal autoregressive integrated moving average exogenous (SARIMA-X) model
• Goodness of fit - Likelihood function (LLF), Akaike information criterion (AICc) and model's diagnosis.
• Forecast and back-testing
• Simulation

ARCH-GARCH analysis

• Basic operators - EWMA/EWV
• GARCH model.
• Exponential GARCH (E-GARCH) model.
• GARCH in the mean (GARCH-M) model.
• Support for Gaussian, Student's t and exponential power distribution (GED) distributed innovations/shocks.
• Goodness of fit - Likelihood function (LLF), Akaike information criterion (AICc) and model's diagnosis

Factor analysis

• Simple linear regression (SLR)
• Multiple linear regression (MLR)
• Principal component analysis (PCA) and principal component regression
• Stepwise regression
• Generalized linear model (GLM)
• Support for the following Canonical link functions:

Distribution	Name	Link function
Normal	Identity
Poisson	Log
Binomial	Logit
Binomial	Probit
Binomial	Complementary log-log

• Support for Goodness of fit: LLF, AIC, Adjusted R-squared, BIC/SIC and Deviance
• Residual diagnosis.

Advanced (combo) models

• Model definition function.
• Mixed model - likelihood function.
• Goodness of fit - LLF, AICc, and model's diagnosis.

Utilities

• Interpolation functions - Flat forward-backward, linear, and cubic spline interpolation.

Compatibility with Microsoft Excel

NumXL's statistical analysis software is compatible with all Excel versions from version 97 to version 2013 (Office 365), and with Windows versions 9x to Windows 8 (32- and 64-bits).

Release history

See also

• EViews—A statistical package for Windows
• gretl—an open source alternative to EViews
• Rcmdr—an open source R-based alternative to SPSS
• OxMetrics—an alternative econometrics package
• Regression Analysis of Time Series (RATS)
• Comparison of statistical packages
• AREMOS

References

External links

• [Official Website]
• NumXL Getting Started
• NumXL Weekly Tips and Hints
• NumXL Release notes
• NumXL Tutorial Videos
• NumXL Community
• NumXL Facebook
• NumXL Twitter