L1 Introduction - Introduction Some years ago I separated...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
Introduction Some years ago I separated from my spouse of 29 years and my friends, concerned about my singular status, got into the habit of arranging the “dinner party”. Low and behold I would turn up to the event to find sitting opposite me someone of the opposite sex in a similar singular state to myself. Clearly it was my responsibility in this arrangement to determine whether or not I liked the person (I invariably did, but of course I could not afford to be too choosy!) and whether or not they liked me. Over the course of dinner I would watch and listen to the person’s reactions to what I had to say. It was an information collection exercise and at some stage I had to arrange what I had observed, synthesize it if you like, into a form that would enable me to decide whether or not she liked me. All I could do was infer whether or not she liked me and of course my inference could have been completely off the mark. Essentially in this problem I faced four possible outcomes and I was by no means indifferent between them. I could infer that she liked me when she really did like me, which of course was a good outcome (we would probably ride off happily into the sunset!). Alternatively I could infer that she did not like me when she really did not like me, which too is a good outcome (we could enjoy the rest of the meal and go our separate ways). Another possibility is that I would decide that she did not like me when she really did (this is bad news and corresponds to an opportunity for a lifetime of happiness foregone!). Finally I could decide that she liked me when really she did not (this too is bad news, the worst outcome in my mind, it was just plain embarrassing!). Consequently I would require more than just “on balance” evidence to convince myself that she liked me in order to avoid the fourth outcome. So what has this got to do with statistics? The answer is…everything! The dinner party was the very essence of the statistical process. Statistics is about dealing with uncertainty, it is about using things we observe and know to learn and understand more about things that we cannot observe, do not know and about which we are uncertain. It is about handling information, deciding what is relevant and what is not, understanding the nature of the probability of events, synthesizing the relevant material and arranging it into a fashion that is useful for making choices in an uncertain environment. Ultimately it is about making those choices in a way that holds the chance of making least preferred mistakes at some acceptable level. The things we know and are going to use come in the form of theory and data. The theory, which will be developed throughout the course, is based upon ideas about the probability of events. It will tell us about the properties of the techniques we are going to use to synthesize the data, it will tell us how to best organize the decision process, it will tell us the nature of uncertainty under certain circumstances and it will tell us how to best
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 2
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 11

L1 Introduction - Introduction Some years ago I separated...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online