For further information, you can get in contact with the experts.

Nowadays we are experimenting a data revolution, which means more capacity for data storage and more data volume.

Also, more variety due to newly created tools and ways of communication such as social media, webs and apps. And finally, data velocity and data processing related to data generation processes and business models constraints.

These three concepts, Volume, Variety and Velocity define the “Big data” term. Therefore, if this definition can be applied and integrated into your business processes, it means that you might have the needs of massive data processing and/or analysis.

Many organizations believe that advanced analytics can potentially enhance their business processes related to logistics, marketing, operations and strategy.

Often, these processes are covered by data modelling techniques and methodologies as the core engine of the decision support systems. The decision support systems allow discovering valuable knowledge from the data that reflects the business status and trends always taking into account historical information.

Therefore, an assisted decision making is much more effective than just expert knowledge and intuition only.

Difference between the parametric and non-parametric modelling.

Now, we can introduce a more technical part. There are many tools and solutions related to “Big data” and advanced analytics. The real-world problems are very complex and it is very important to choose an appropriate technique or it even might be the case in which one would need to design their own custom methodology.

There exist two main types of models: parametric models and non-parametric models.

In a parametric model, the data follows a probability distribution with a fixed set of parameters. This is a very appealing setting because the initial assumptions of the model are fulfilled and it is well-based from the theoretical point of view.

The problem with it might arise, for example, when one has multimodality behaviour in the data, which means that many models such as Quadratic Discriminant Analysis (QDA) or Linear Discriminant Analysis (LDA) (Hastie et al. 2008) are not suitable anymore and that the inferences one does might be completely misleading. But for instance, Fisher Linear Discriminant Analysis (FLDA) (Hastie et al. 2008) doesn’t need the normality assumption, neither Principal Component Analysis (PCA) (Jolliffe, I.T. 2013).

 The non-parametric models can be regarded as:

• distribution-free methods and
• methods without predefined structure. First sub-type would include models such as kNN, Decision Trees, PCA and the second sub-type would include Support Vector Machines (SVM) (Bishop 2006) and Non-parametric Local Polynomial Regression (Tibshirani 2017).

For instance, in SVM it is very clear that the number of parameters grows with the amount of data as the number of support vectors is bounded by the number of observations. And in contrast to parametric regression models with linear predictor function in which the relationship between the independent and dependent variables is predefined, non-parametric regression provides a more flexible framework as the linearity assumption and the form of predictor function are relaxed, however the errors are still i.i.d. Note that the linearity is always with respect to the parameters, not the data. This will be illustrated in the following example.

In the example from (Pagan and Ullah, 1999) it is taken the logarithm of wage in terms of the age of Canadian male individuals with common education level in 1971. The relationship is assumed to be quadratic in age. The formula of the parametric linear model would be as follows:

Note that this model is still linear with respect to the parameters.

In Fig. 2 it is shown the parametric regression fit versus the non-parametric model.

For the non-parametric model, the bandwidth parameter is estimated by cross-validation using AIC criterion. Actually, in non-parametric modelling, as opposite to linear parametric models, often it is convenient to not to assume homoscedasticity and estimate the conditional variance. The following R code has been used to generate the presented example:

library(“np”)

data(“cps71”)

model.par <- lm(logwage ~ age + I(age^2), data = cps71)

summary(model.par)

model.np <- npreg(logwage ~ age, regtype = “ll”, bwmethod = “cv.aic”, gradients = TRUE, data = cps71)

summary(model.np)

plot(cps71$age, cps71$logwage, type=”o”, ylab=”logwage”, xlab=”age”, main=”Linear Model vs Local Polynomial Regression”)

lines(cps71$age, model.par$fitted.values, col=”lightblue”, lwd=2)

lines(cps71$age, fitted(model.np), col=”salmon”, lwd=2)

legend(“topleft”, legend = c(expression(paste(“LM (“, R^2, “=0.23)”)), expression(paste(“Local-PR (“, R^2, “=0.33)”))), col=c(“lightblue”, “salmon”), lwd=2)

La entrada The parametric and non-parametric modelling for Big Data se publicó primero en .

La entrada Ventajas del modelo de optimización con RPA y AZURE se publicó primero en .

La entrada Blockchain, ¿cómo afecta la Seguridad y privacidad de una empresa? se publicó primero en .

La entrada ¿Porqué migrar a SAP S/4 Hana con Besh? se publicó primero en .

La entrada “Súbete a la Nube” – Compartimos contigo nuestra experiencia en Madrid Fly se publicó primero en .

La entrada Estuvimos en el AUSAPE 2017 y queremos compartirlo se publicó primero en .

La entrada Alianza con Huawei Empresas para proyectos sustentados en tecnología SAP se publicó primero en .

Cloud Computing is a concept that has been installed a few years ago in the business sector as well as in the digital economy and the SAP HANA Cloud Platform has a lot to do with this. One of SAP’s services for IoT (Internet of Things) is the SAP HANA Cloud Platform, a service that provides the ability to quickly manage, deploy and develop applications that synchronize data in real time, from one machine to another, As well as remote data, which allows you to manage your entire business in real time and perform operations in a smart and fast way. This also improves decision making in the perimeter of the network and optimizes business processes in the center of the organization.

The SAP HANA Cloud Platform is a cloud platform of memory technology, designed for everyone who is increasingly online and data based. With this platform, you can create applications, taking advantage of the speed and scales of SAP HANA.

Many companies are developing this platform for their business, because it implements easy to use applications on any device and, and of course, connects the company to the digital economy. SAP HANA Cloud Platform has a technical variety to build applications on a single platform, taking advantage of the capabilities of SAP HANA along with different business services.

CHARACTERISTICS

SAP HANA Cloud Platform enables customers and partners to quickly integrate, create, and manage cloud-based enterprise applications, complementing them to SAP solutions. You can deploy and manage connected remote devices. Provides access to a feature-rich, easy-to-use cloud environment. The data center infrastructure is reliable and secure. It is the best connector of your company with the digital economy. The data is analyzed in real time and in an advanced way, directed to the benefit of the client.

In short, it is not only an in-memory technology database, but a complete application platform itself. Within the new technologies, there is no doubt that SAP HANA Cloud Platform is a bet on the future within the innovations for the digital economy, which increasingly needs these tools to accelerate its performance, in a world that grows exponentially in its Volume of information, within the Cloud Computing environment.

Source:

SAP Blog

La entrada SAP HANA Cloud Platform: Innovation in the digital economy se publicó primero en .

The reason for this shift being the excessive amount of information and the digital tools that exist in order to manage all this information. To respond to this demand, new professional profiles have emerged such as CDO (Chief Data Officer). CDO´s are responsible for analyzing and managing a company´s data to convert them into a business asset. In short, a CDO is resposible for the management of all business data, and the usefulness of this data.

The presence of CDO´s in companies grows every year

According to experts, CDO´s must meet three objectives; Improve customer interaction, increase efficiency and differentiate from its competitors. To meet these goals, these executives must harmonize two different sectors: analytics and data management. To do this, they must combine all the internal and external data of the organization, as well as apply effective analytical and statistical techniques to reach new solutions.

This mix of data management and analytics is becoming the dominant trend in CDO agendas for large companies. Companies should hire this digital profile if they wish to incorporate new models of digital strategies. One of the tasks of the CDO´s is to integrate the data and its analysis in a joint way and not as separate operations.

What does analytics and data management entail?

Data analytics is the science that examines the raw data in order to draw profits and conclusions. For example, decision-making is improved by better knowledge of customers and the market through analytics. In order to obtain patterns and useful data information, analytics applies statistical techniques.

In contrast, data management is the processing and management of data using digital technologies such as Big Data or databases. It is common to use software such as SAP HANA Cloud Platform for data management. The difference with respect to a few years ago, was that companies could not analyze their processes without them being treated accurately. In order to carry out the analysis and management of large volumes of information, CDO´s use softwares such as SAP Hana Cloud Platform. Organizations must build their business in analytics and data management, but they must be built correctly and this is where CDO´s must step in.

La entrada What are the differences between analytics and data management? se publicó primero en .

La entrada Obtenemos el certificado en “SAP Outsourcing Operations Poviders” – Besh se publicó primero en .