CourseTo equip students to use basic statistical techniques for exploratory data analysis. To present problems in the field of logistics that involve using statistical data analysis in decision-making.Statistics books for the studentBUSSAB, Wilton de O.; MORETTIN, Pedro A. Estatística básica (Basic Statistics). Saraiva, 2017.Attendance will be taken! There will be a quiz and an exam!IntroductionWhat is statistics?Statistics is a set of methods used to analyze data;It can be applied in practically every area of human knowledge, through observation and inference based on data;Its goal is the collection, reduction, analysis and modeling of data;These are methods that let us answer questions scientifically, with as little effort as possible.Statistics contributes to every step of the scientific method. In short, in this method the scientist seeks to answer a question and often to test a hypothesis; they collect information from experiments or observations and draw conclusions from it.Statistics helps through methods of data visualization and analysis, which extract the necessary information from the data.A researcher wants to know whether a new drug lowers blood pressure. At which stage of the scientific method does statistics come in?Answer: In collecting the data, analyzing it and drawing conclusions from it — statistics decides how to collect (how many patients, how to randomize them), summarizes and analyzes the measurements, and says whether the observed difference is large enough to conclude something. It is not confined to the end of the work.DataData... data... data... in the end, isn't everything data? Let's analyze the following statement:“Today it is hot”- Some guy on the streetDepending on your point of view, "today it is hot" can be a piece of information or just a piece of data. If we classify days only as hot or cold, "hot" is a datum that, on its own, carries little relevant information: it could be 25 or 38 degrees Celsius, we do not know which temperature separates a hot day from a cold one (nor whether we are talking to an Inuit or to someone from Rio de Janeiro). In another situation, if we see a person sweating while saying the sentence above, we use prior knowledge and realize that "hot" means an environment uncomfortable for the human body; with that, we have information: the temperature is probably above 30 degrees Celsius.We see, then, that an isolated datum, without context, has little meaning and leads to no understanding. On its own, it cannot support conclusions, let alone decisions.Information, on the other hand, is the result of ordering and organizing data, within a context, so that they can be interpreted and convey meaning.Finally, much like in the scientific method, from information we generate knowledge about the situation at hand.We can thus build a pyramid in which the farther we move from the data, the more abstraction we have [1]:With data we can generate information and, later, knowledge — in the figure: Dado = data, Informação = information, Conhecimento = knowledgeAnother example to understand abstraction:MANGA.Can we get any information from the datum "manga"? In Portuguese, "manga" means both the fruit (mango) and the sleeve of a shirt. Which one is it? To generate information, we need the context in which the datum was found, or some level of abstraction!If we carry this idea over to other fields, we see that all of them are subject to it. Presenting information correctly, with no room for other interpretations, is very hard, and that concern is also part of statistics.The "today it is hot" problem raises other statistical concerns: how do we classify a day as hot or cold? Which temperature separates one from the other? Why did we say "probably above 30 degrees"? Later on, we will see that we do not need to observe every day to draw conclusions about the climate of a place: with a well-chosen sample and with probability, we can estimate what happens in the whole.Note that, in a world with infinite time and money, statistics would be much simpler and this discussion would matter less. If in doubt about the data, we would just analyze all of it; not knowing what "hot" means, we would just go back and measure every day. The problem is that the real world does not work like that and, without statistics, simple questions would be very expensive to answer.A spreadsheet has a column with the values 37.8; 38.5; 36.9, with no heading and no unit. In terms of the data, information and knowledge pyramid, what is this column?Answer: Data, because without context we do not know what they mean — only when we know that they are, for example, the body temperatures in degrees Celsius of three patients do the numbers become information. Being organized in a spreadsheet is not enough: what turns data into information is context and interpretation.Why does statistics work with samples instead of always observing every case?Answer: Because observing every case is usually too expensive or too slow, and a well-chosen sample lets us estimate the whole — that is the idea of the world without infinite time and money. A sample does not give an exact result: it gives an estimate, and probability measures its margin of error.Something to think about1) What data do companies like Facebook and Google collect?2) What information can we extract from it?ConclusionTherefore, the usefulness of statistics is to:Reduce the cost of the scientific method in time and money;Convey information as clearly as possible, avoiding misinterpretation;Generate knowledge and conclusions from data analysis (statistical inference);Create standardized, scientific methods to answer questions.A shop records that it sold 120 umbrellas on a Monday. Which option already represents knowledge, and not just data or information?Answer: The conclusion that umbrella sales go up on rainy days, which lets the shop increase its stock when rain is forecast — the number alone is data; the sentence and the table, with context, are information; relating sales to the weather and using it to make decisions is knowledge.References[1] VALENTIM, Marta Lígia Pomim. Inteligência competitiva em organizações: dado, informação e conhecimento (Competitive intelligence in organizations: data, information and knowledge). DataGramaZero, Rio de Janeiro, v. 3, n. 4, Aug. 2002.
ردود الفعل
0
0 تعليقات
كن أول من يعلق
لا شيء هنا بعد.