Software:NumXL

NumXL

Developer(s)	Spider Financial Corp
Initial release	April 15, 2009 (2009-04-15)
Stable release	1.64.42788.1 / December 25, 2016 (2016-12-25)
Written in	C, C++
Operating system	Windows
Type	Econometrics software
License	Trialware
Website	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.
• Periodogram

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	[math]\displaystyle{ \mathbf{X}\boldsymbol{\beta}=\mu\,\! }[/math]
Poisson	Log	[math]\displaystyle{ \mathbf{X}\boldsymbol{\beta}=\ln{\left(\frac{\mu}{1-\mu}\right)}\,\! }[/math]
Binomial	Logit	[math]\displaystyle{ \mathbf{X}\boldsymbol{\beta}=\ln{\left(\frac{\mu}{1-\mu}\right)}\,\! }[/math]
Binomial	Probit	[math]\displaystyle{ \mathbf{X}\boldsymbol{\beta}=\Phi^{-1}{\left(\mu\right)}\,\! }[/math]
Binomial	Complementary log-log	[math]\displaystyle{ \mathbf{X}\boldsymbol{\beta}=\ln{\left(-\ln{\left(1-\mu\right)}\right)}\,\! }[/math]

• 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

Version	Release	Year	Release Date	Notes
NumXL 1.0	1.0	2009	October 1, 2009	Official release
NumXL 1.5	1.5	2011	June 13, 2011	New version
NumXL 1.51	ORB		September 23, 2011	New version
NumXL 1.52	ORB	2012	January 26, 2012	New version
NumXL 1.53.41023.1	SAC		April 25, 2012	New version
NumXL 1.54.41030.2	RED		May 2, 2012	New version
NumXL 1.55.41040.2	LYNX		May 11, 2012	New version
NumXL 1.56.41060.2	ZEBRA		May 31, 2012	New version
NumXL 1.57.41168.2	SINGA		September 17, 2012	New version
NumXL 1.58.41197.1	BAJA		October 16, 2012	New version
NumXL 1.59.41245.1	TUCSON		December 3, 2012	New version
NumXL 1.60.41413.1	APACHE	2013	May 20, 2013	New version
NumXL 1.61.41445.1	BOLAS		June 20, 2013	New version
NumXL 1.62.41788.1	DEWDROP		Nov 24, 2013	New version
NumXL 1.63.41810.1	SHAMROCK	2014	June 6, 2014	New version
NumXL 1.64.42727.1	TURRET	2016	December 25, 2016	New version
NumXL 1.65.42877.1	HAMMOCK	2017	May 19, 2017	New version

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

External links

• NumXL Getting Started
• NumXL Weekly Tips and Hints
• NumXL Release notes
• NumXL Tutorial Videos
• NumXL Community
• NumXL Facebook
• NumXL Twitter