Search results

Proposes cusum‐cuscore procedures that consider equal weight past sample mean values and past sample score values. The objective is to compare the performance of the cusum‐cuscore and the ewma‐cuscore and in particular to investigate the number of past sample mean values needed to make a significant impact on the performance of the schemes. It will be shown by average run length calculations that the two schemes complement each other very well and that they perform significantly better than standard Shewhart, cuscore and cusum procedures.

The proposed exponentially weighted moving average combined Shewhart cumulative score (EWMA‐CUSCORE) procedure for controlling the process mean cumulate scores of ‐1, 0, 1 or 2h assigned to each moving average of the current and past sample mean values depending on a preassigned interval in which its value falls. It will be shown by average run length (ARL) comparisons that the proposed scheme performs better than the Shewhart type schemes, the combined Shewhart cumulative score type schemes, the cusum type schemes and the standard EWMA type schemes for detecting shifts in the process mean when the underlying process control variable is normal.

Combined Shewhart‐cumulative score (cuscore) quality control schemes are available for controlling the mean of a continuous production process. In many industrial applications, it is important to control the process variability as well. The proposed combined Shewhart‐cumulative‐score (cuscore) procedure for detecting shifts in process variability uses the procedures developed by Ncube and Woodall (1984) to monitor shifts in the process mean of continuous production processes. It is shown, in the one‐sided case, by average run length comparisons, that the proposed schemes perform significantly better than comparative Shewhart procedures and in some cases even better than cusum schemes when using some process variability quality characteristics.

The purpose of this research is to provide companies and consumers with information about the potential connections between data breach types and institutions. This study also aims to add to the body of knowledge about data breaches.

This study analyzes a chronology of five years of data breaches. The data were classified and analyzed by breach and institution type, record size, and state. Multiple statistical tests were performed.

Breach types stolen and exposed are statistically more likely to occur. Educational institutions are more likely to have a breach and it is more probable that educational breaches will be of type hacker or exposed. The proportion of insider incidents is smaller than the other breach types. The number of records breached is independent of institution and breach type.

Only those breaches with a specified number of records are included. The information used may have been updated after our analysis, usually a change in the number of records identified.

Additional knowledge about characteristics of data breaches and the relationship between breach types and institution types will enable both businesses and consumers to be more effective in protecting sensitive information. Businesses will be able to create security budgets based on risk factors and consumers will be more aware of the risks of providing sensitive information.

This study provides a longitudinal analysis covering five years of data breaches and analyzes the relationship between five breach types and six types of institutions.