Through the ExMetrix platform it is possible to build a model of a process or a phenomena that occur around us.
There are two modes of modelling that we can choose from:
1. Simple Mode – it is available from the main menu and building a forecasting model is almost immediate. It is for people who take first steps in data analytics. The user enters just a few parameters. The properties left are set by the system automatically.
2. Advanced Mode – it is mainly dedicated to users who have the analytics and modelling knowledge . This time the user defines parameters and mathematical form of the model.
The following guide presents the Easy Mode. The process starts after choosing the option „Make prognosis” from the main menu.
Next we determine a desired variable – we name it a reference variable or a dependent variable. In the Simple Mode we choose the reference variable from the user’s profile.
The profiling aims at easier access to variables without a necessity of searching for it in a huge database. The prediction of any reference variable – that we can’t find in profiles – is possible in the Advanced Mode. In the following example we assume that the user belongs to a category „investor” and is interested in prediction of a value of CHF/PLN currency pair relation (The Swiss franc to the Polish zloty). In this case the user chooses the “investor profile” and selects a currency.
The user receives on-screen information about available currencies. Then the user selects the variable (in this case CHF/PLN currency pair).
The user gets an on-screen information about search parameters.
The parameter „Frequency” determines a time interval. We choose the proper time interval from a drop down list depending on whether we search for daily, monthly or other predictions.
The system automatically checks a suitability of all variables in the context of prediction model inputs. Besides, the system converts all variables to a quotation interval selected by the user. If the users chooses, e.g. the monthly interval it causes that daily, quarterly or annual data will be converted to the monthly data.
The “Model prognosis” option requires setting of time horizon of forecasts. The time horizon is expressed in time units that were set in “Frequency”. If we choose a month in “Frequency” box and we set Model Prognosis of a value of 5, we will build models that forecast 5 months forward. If the frequency is expresses in days and Model Prognosis is defined as 5, we will receive the model that forecasts 5 days forward.
After setting the parameters of process we confirm a job by “Make prognosis” icon.
The system confirms the beginning of the process by displaying an on-screen information.
After the process we receive ready and optimised forecasting model.
The estimated model:
It is possible to re-optimise the estimated model (without random choice of initial parameters) or we can also start the process again, choosing the initial parameters with unchanged remaining parameters.
After the database update, the model response can be upgraded (computed on a basis of before estimated values of structural parameters).
It is also possible to use estimated parameters of model and its answers to individual changes of given variables (we can check if the model is susceptible to change of variables or not and whether it may be the basis of reduction of its value eliminating variables of weak influence. Then we can estimate the model again).
File to download: