Scenario analysis and sensitivity analysis.
Over-investing and under-investment
Over-investing refers to the process and practice in which an individual chooses to make investments on an asset which are more than the worth of the asset in the open market. This kind of investment is mainly seen for people who make personal and expensive consumable investments such as trailers, automobiles, and houses. For example, if a house owner new renovations and addition to his home to the situation that he uses extra spending than exceeds the market of other houses, then the situation is referred to as overinvestment. As a result, the neighbourhood effect will devalue the home to the extent that it becomes of less worth as compared to the investment cost.
On the other hand, underinvestment is referred to a state in which an individual invests less money in an asset than what is needed. In under-investment, shareholders of a company choose not to invest in low-risk deals to provide safe cash flow for the benefits of shareholders. They instead decide to make a high-risk investment that will yield high-profit assets that will help to enhance their share value. Underinvestment brings up a problem which is commonly referred to as underinvestment problem. Underinvestment problem usually arises between debt holders and the shareholders. The challenge makes the company forgo investments opportunities that are more valuable because the debt holders are likely to gain the benefits and leaving shareholders with fewer returns.
Censored and truncated data
These two types of bounded data that have direct implications in the manner we work with them in the analysis (Klein and Moeschberger). Edited usually contain unknown values after the bound any side of the number line or in both. Censored data exists in various ways such as be design, and so on. Therefore, when data is observed, it is reported at the end of the line. The researcher then restricts the range by making various decisions. For example, recording pollutants in water is an example of a lower censoring boundary. In this case, the researcher fails to detect the level of these pollutants, especially when they fall below a given threshold. Therefore, the pollutants that fall under the set threshold are not recorded. Also, an excellent example of the upper sensor is the one that is put on temperatures during the science experiment. In this case, when the temperatures exceed the upper sensor, it has no impact. Another excellent example of censored data is when income is reported as being less than or more than a given amount. Lastly, test scores and admission tests are also examples of censored data. In schools, for instance, students cannot score more than 100% regardless of how smart they are.
Conversely, truncated data is the one that occurs when values are excluded when analysed or gathered. For example, police make several arrests and then one police make zero arrests, then he will be excluded from the sample. Another example is for a group that limits the entry of membership based on the age; new members should not be less than 18 years or above 35 years. This kind of data usually happens when boundaries are set to limit the cases that do not meet the requirements.
The differences between censored data and truncated data is that; censored data occurs when the boundary must not be exceeded because the values below the boundary don’t count, while in truncated data, the data set excludes the observations of the data collected above or below the boundary.
Work Cited
Klein, John P., and Melvin L. Moeschberger. Survival analysis: techniques for censored and truncated data. Springer Science & Business Media, 2006.