NeuroXL Predictor

  • 99.95$
    Buy Now Free Trial

Compatible with Windows 2000, XP, Vista, 7, 8, 10 and Microsoft Excel 2000-2016


Table of Contents


The Value of Neural Network Forecasting and Estimation

Estimating and forecasting future conditions govern many critical business activities, such as inventory control, procurement of supplies, labor cost estimation, and prediction of product demand. Inaccurate or misleading estimates can result in aggravation or total company chaos. Lost sales, inefficient allocation of resources, and bloated inventories are often a direct result of forecasts that failed to predict future conditions. In the financial industry, inaccurate forecasts for stocks and other investments can result in poor trades and lost opportunities for gain. Inaccurate predictions of sporting outcomes can result in poor wagering and resulting losses.

Microsoft Excel provides a built-in tool for predictions, but the accuracy of its results is significantly reduced when non-linear relationships or missing data are present, which is often the case when analyzing historical business, investment, or sports data. Neural networks are a proven, widely used technology for such complex prediction problems. Loosely modeled after the human brain, neural networks are interconnected networks of independent processors that by changing their connections (known as training) learn the solution to a problem.

One of the main reasons analysts have been slow to use advanced methods such as neural networks to improve forecasts is that such methods can be difficult to master. NeuroXL Predictor removes the psychological and practical barriers by hiding the complexity of its advanced neural network-based methods while taking advantage of analysts' existing knowledge of Microsoft Excel spreadsheets. Since users make predictions through the familiar Excel interface, learning time is minimal, greatly reducing the interval between loading the software and performing useful predictions. The application is extremely intuitive and easy-to-use for beginners yet powerful enough for the most demanding professionals. Also, using NeuroXL Predictor requires no prior knowledge of neural networks.

In summary, these are the key advantages of NeuroXL Predictor:

>Easy to learn and use
>No prior knowledge of neural networks required
>Integrates seamlessly with Microsoft Excel
>Provides proven neural network technology for highly accurate forecasts
>Lowest cost neural network product on the market

Prime applications for NeuroXL Predictor include

Forecasting Stocks and other Investments: The ability of NeuroXL Predictor to discover non-linear relationships in input data makes it ideally suited for forecasting dynamic systems like the stock market. The price of stocks and other investment vehicles such as bonds, derivatives and options are also influenced by many different factors that are often interrelated. Traditional forecasting methods, such as regression and data reduction models, are limited in their effectiveness as they make assumptions about the distribution of the underlying data, and often fail to recognize the interrelatedness of variables. NeuroXL Predictor, drawing on the latest in artificial intelligence research, recognizes even subtle relationships between variables.

In addition to stock market prediction, NeuroXL Predictor is also ideally suited to making predictions in other financial areas, such as:

>Foreign exchange trading
>Financial planning
>Commodity trading
>Economic forecasting
>Currency trading
>Corporate bond ratings
>Oil and gas trading

Sales Forecasting:Sales forecasting lets businesses make better purchasing decisions and manage inventory more efficiently. It can also identify new opportunities for increased sales. However, sales forecasting is extremely difficult, since a large number of factors influence sales that are often interrelated. Price, seasonality, advertising, and competitor behavior are all factors that influence sales. NeuroXL Predictor analyzes historical data to learn the interrelation between these factors, and makes predictions of sales levels for various combinations.

In addition to sales forecasting, NeuroXL Predictor is ideal for many other business activities, including:

>Industrial process optimization
>Loan approvals
>Credit scoring
>Marketing campaign prediction
>Cost prediction

Sports Predictions: NeuroXL Predictor's ability to handle multiple non-linear relationships makes it ideally suited to predicting the outcome of team sports and racing events. For example, with thoroughbred horse racing, NeuroXL Predictor can simultaneously examine the relationship between multiple contenders, resulting in more accurate predictions of winners than traditional estimation techniques. A typical horse or greyhound race involves a complex problem domain, often with 50 performance variables for 6-9 animals. NeuroXL Predictor's robust neural-network-based implementation allows it to detect relevant patterns in such data sets, resulting in high-quality predictions.

Back to Contents

The Advantages of NeuroXL Predictor

Most neural network prediction software requires the user to learn about neural networks, complete large tutorials, and/or perform data pre-processing. NeuroXL Predictor requires no prior knowledge of neural networks and hides the complexity from the user, allowing for useful predictions often just minutes after installation. Users just need to specify input and output references, perform a few mouse movements, and their prediction is returned.

NeuroXL Predictor also offers compatibility with Microsoft Excel-based trading software and with the entire AnalyzerXL product suite.

Back to Contents

Example: Using NeuroXL Predictor to Predict Stock Prices

Since NeuroXL is an add-on to Microsoft Excel, it can perform predictions on your existing historical data and technical functions already in spreadsheet form. Let's say you have a table with historical stock price data and two technical analysis functions, and you wish to do a prediction of tomorrow's closing price.

The steps to predict tomorrow's closing price are:

1. Select NeuroXL Predictor from the menu in MS Excel.

This will launch the program, bringing up the NeuroXL Predictor dialog box.

 2. Using the mouse button choose the range of numerical data (Inputs) of the table that you want to use and which is influencing on the outputs and choose the range for the Outputs. We set the range of inputs to be all the data from B13 to H84. The outputs are specified as the values from E14 to E85.

3. Align the range if necessary. Check "Scale input and output values" box if you would like the values of the range specified to be scaled. Neural networks require all input and output data to be in the allowable range of the activation function. This option scales the data to this range and when outputting the result the data is being scaled back. You can also set the number of epochs, minimum weight delta and choose whether to view the learning process.

Epochs. Epoch is a full cycle of neural network training on the entire training set. This parameter defines the maximum learning epochs cycles to reach specified minimum weights delta.

Minimum weight delta. Within the training process synapses weights are being corrected. Minimum weight delta defines the desired lowermost weight correction value when the network is considered to be trained enough.

4. Set the following parameters (for advanced users):

>Initial weights: Initial weights of synapses. Synapses of every neuron will be initialized with random values from 0 to the initial weights.
>Learning rate: a value between 0 and 1 that affects the rate at which the network learns. The larger the learning rate, the faster the network will converge. Be advised that oscillation and non-convergence may occur if the learning rate is set too high.
>Momentum: High learning rates often lead to weight change oscillations during the training process, which may cause non-convergence or return of a non-optimal solution. Momentum makes it less likely for such undesirable cases to occur by making the next weight change a function of the previous weight change to provide a smoothing effect. The value for momentum (between 0 and 1) determines the proportion of the last weight change that is added to the next weight change.
>Activation function: There are five functions available: Threshold, Hyperbolic tangent, Zero-based log-sigmoid, Log-sigmoid and Bipolar sigmoid.
>Neurons in hidden layer: Set the number of neurons in hidden layer

Note: In most cases, the default values are acceptable for these parameters.

5. Click the New button, followed by the Start training button. If "Show learning process" option is checked, a graphical representation of the training process is displayed. The green graph shows the actual values, and the red the predicted ones:

6. When the training is complete, you should specify the input and output ranges for the prediction and press Predict button. The result is in E87 cell.

In six steps, we have performed a complex neural network prediction, taking into account non-linear variables and the results of two technical analysis functions. NeuroXL Predictor does the work of determining the relationship between variables and detecting relevant patterns in the data. All the user needs to do is specify the inputs and outputs and set the required parameters.

Back to Contents


NeuroXL Predictor is a powerful, easy-to-use and affordable solution for advanced estimation and forecasting. By harnessing the latest advances in artificial intelligence and neural network technology, it delivers accurate and fast predictions for your business, financial, or sports forecasting tasks. Designed as an add-on to Microsoft Excel, it is easy to learn and use yet powerful enough for the most demanding professional.

Back to Contents



New versions of NeuroXL Predictor, NeuroXL Clusterizer and NeuroXL Package released: 4.0.6

July 18, 2016


Subscribe to news


I can definitely recommend NeuroXL software to any individual or business that would like to take advantage of the power of artificial neural networks in analyzing complex data.

Dr. Jean-Michel Jaquet