3. Data Preparation Phase
Fifty-six financial ratios were identified by the researchers through a search of the literature on bankruptcy prediction, 16 of which were then dropped due to duplication. There remained 40 financial ratios in the data set, including measures of growth, profitability, safety/leverage, activity/efficiency, and productivity. (more…)
Most short-term investment issues (other than U.S. governments) are rated by one of the credit rating agencies, for example, Moody’s or Standard & Poor’s. The rating reflects the creditworthiness of the issuer, the amount being raised given existing debt, collateral or backup credit facilities, and the general level of interest rates. Besides the issuer default risk, a number of other factors influence the overall rate or yield on an instrument. When deciding between various investment options, the cash manager needs to convert all of the rate information to an annualized yield so that the investment options may be compared on a common basis. (more…)
1. Business/Research Understanding Phase
The recent economic crisis in East Asia has spawned an unprecedented level of corporate bankruptcies in that region and around the world. The goal of the researchers, Tae Kyung Sung from Kyonggi University, Namsik Chang from the University of Seoul, and Gunhee Lee of Sogang University, Korea, is to develop models for predicting corporate bankruptcies that maximize the interpretability of the results. They felt that interpretability was important because a negative bankruptcy prediction can itself have a devastating impact on a financial institution, so that firms that are predicted to go bankrupt demand strong and logical reasoning. (more…)
Fistlike Gestures. Dr. Louis Loeb, in a paper entitled The Fist, finds that this gesture communicates from the unconscious to the unconscious. His premise is that one can affect anothers reaction by merely clenching the hand in a fistlike gesture and using it to emphasize the verbal language. People who clench their fists may allow them to be clearly seen, but more often they conceal the gesture by thrusting the fist in their pocket, tucking both fists under the armpits in a crossed-arms gesture, or putting both hands behind the back. Fist-clenching is essentially a masculine gesture. It is somewhat unusual to see a woman making a fist while talking. (more…)
You can make any cell active by clicking it with the mouse. You can also move the active cell using the arrow, page up, and page down keys. If you try to move off the screen, the worksheet scrolls to keep the active cell in view. Excel also defines a few special keystrokes for moving the active cell: (more…)
4. Modeling Phase
Rules were restricted to having only a single antecedent and a single consequent. Many interesting association rules were uncovered, along with many uninteresting rules, which is the typical scenario for association rule mining. One such interesting rule was: If place of birth = Vietnam, then law type = criminal law, with 90% confidence. (more…)
This example demonstrates a very important principle: In a market dominated by risk-averse investors, riskier securities must have higher expected returns, as estimated by the marginal investor, than less risky securities. If this situation does not exist, buying and selling in the market will force it to occur.We will consider the question of how much higher the returns on risky securities must be later in the chapter, after we see how diversi?cation affects the way risk should be measured. Then, in Chapters 4 and 5, we will see how risk-adjusted rates of return affect the prices investors are willing to pay for different securities. (more…)
1. Business/Research Understanding Phase
The researchers, Sasha Ivkovic and John Yearwood of the University of Ballarat, and Andrew Stranieri of La Trobe University, Australia, are interested in whether interesting and actionable association rules can be uncovered in a large data set containing information on applicants for government-funded legal aid in Australia. Because most legal data is not structured in a manner easily suited to most data mining techniques, application of knowledge discovery methods to legal data has not developed as quickly as in other areas. The researchers goal is to improve the delivery of legal services and just outcomes in law, through improved use of available legal data. (more…)
While you can easily learn how to organize yourself better and manage your time better, you can’t always get rid of the pressure and stress you experience each day. You can, however, learn how to channel that energy and make yourself more productive, as opposed to being too frazzled to think clearly. (more…)
4. Modeling Phase
The data were split into a training set (80% of the data) and a validation set (20%). A neural network model was applied, which uncovered the following results:
a. Certain industries had the most predictable abnormal stock returns, including:
