Questions and Answers of Investments Analysis And Management

The data show the chemical composition for nine oxides of 48 specimens of Romano-British pottery determined by atomic absorption spectra (for more information on the analysis of compositional data,
In the SID data by coding a variable, group, as controls = 1 and SID victims = −1, show the equivalence of a multiple regression model for group as the response variable and explanatory variables
For the data from the risk-taking investigation, produce a scatterplot of “High” versus “Low” scores showing the three groups. As an exercise, construct discriminant functions for each pair
Return to the body measurement data introduced in Chapter 12, and find Fisher’s linear discriminant function for allocating individuals to be men and women. (We are aware that there is a foolproof
Repeat the analysis of the life expectancy data for the corresponding data for women.
The data give the protein consumption in 25 European countries for 9 food groups. Is there any evidence that the countries cluster in some way?
Apply the model-based clustering approach to the data on life expectancies and compare the results with those from the k-means clustering given in the text. Do the same for the crime rate data.
The data give the lowest temperatures in degrees Fahrenheit recorded in various months for cities in the United States. Plot the data in any way that you think might be helpful and explore whether
Apply k-means to the crime rate data after standardizing each variable by its standard deviation. Compare the results with those given in the text found by standardizing by a variable’s range.
Reanalyze the data on life expectancies standardizing the variables.How do the results compare with those given in the text for the unstandardized data?
The following matrix gives the correlations between ratings on pain made by 123 people suffering from extreme pain. The nine statements are 1. Whether or not I am in pain in the future depends on the
Investigate the use of alternative rotation methods to varimax on the crime rate data.
Apply principal factor analysis to the crime rate data, and compare the varimax-rotated solution to that given in the text and found using the maximum likelihood estimation. Also compare the
Returning to the small exploratory factor analysis example described in Section 15.2, suppose now that the observed correlations had beenFind the values of the parameters in a one-factor model fitted
Table 14.12 gives data on the cross-classification of people in Caithness, Scotland, by eye and hair color (Fisher 1940). The region of the UK is particularly interesting as there is a mixture of
The elements of the matrix in Table 14.11 result from averaging the ratings of 18 students on the degree of similarity between 12 nations on a scale ranging from 1, indicating “very different”,
The elements of Table 14.10 give the Mahalanobis distances (see Chapter 12 for a definition) between ten types of galaxy calculated from raw data on seven variables such as diameter, brightness, and
to determine the number of dimensions required to represent these distances. Construct an appropriate plot (or plots) to visualize the solution with this number of dimensions.
Return to the data on road distances in Finland and now use the fit criteria defined in Technical Section
Rescale the coefficients defining the principal components of the crime rate data so that they represent correlations between the components and crime rates.
The data give prestige, income, education, and suicide rates for 36 occupations, originally given in Labovitz (1970). Undertake a PCA of the data and use the results to try to answer the question of
Macdonnell (1902) obtained measurements on seven physical characteristics for each of 300 criminals. The seven variables measured were (1) head length, (2) head breadth, (3) face breadth, (4) left
Find the principal components of the following correlation matrix and compare how the one- and two-component solutions reproduce the matrix. 1.0000 0.6579 1.0000 R = 0.0034 -0.0738 1.0000
The crime rate data considered in the text contains a number of possible outliers. Reanalyze the data using principal components after removing the observations you consider to be outliers, and
The data give the life expectancies in different countries by age and by sex. Find numerical summaries separately for men and women, and construct suitable graphics for an initial examination of the
For the paint-sprayer data, use some suitable graphics to identify any observations that you think are outliers. Construct chi-square plots of the generalized distances after removing the outlying
For the body measurements data, using the scatterplot matrix of the data as a guide, try to identify the men and the women in the sample and then find the means, variances, and covariance and
The data in Table 11.10 given in Davis (2002) arise from the Iowa Cochlear Implant Project to compare the effectiveness of two types of cochlear implants in profoundly and bilaterally deaf patients.
With the same data, find the correlation matrix of the measured length and the three estimated lengths using 1. The complete case approach (i.e., the listwise deletion)2. Mean value imputation 3.
The data in Table 11.9 show subjective estimates of the lengths (to the nearest 0.1 inch) of 15 pieces of string as assessed by three raters. Also given is the accurately measures length of the
The data shown in Table 10,13 were collected in a follow-up study of women patients with schizophrenia (Davis, 2002). The binary response recorded at 0, 2, 6, 8, and 10 months after hospitalization
Investigate the use of other correlational structures than the independence and exchangeable structures used in the text, for both the respiratory and the epilepsy data.
For the epilepsy data investigate what Poisson models are most suitable when subject 49 is excluded from the analysis.
The data give the plasma inorganic phosphate levels for 33 subjects, 20 of whom are controls and 13 of whom have been classified as obese (Davis, 2002). Produce separate plots of the profiles of the
The data arise from a trial of estrogen patches in the treatment of postnatal depression. Women who had suffered an episode of postnatal depression were randomly allocated to two groups: the members
Investigate whether there is any evidence of a treatment × time interaction in the BtB data.
For the BtB data, construct a plot that shows the mean profiles over time for each treatment group and has appropriate error bars at each time point.
Five different types of electrodes were applied to the arms of 16 subjects and resistance measured in kilohms (the first eight subjects are women and the remaining eight men). The resulting data are
For the pain score data replace each missing value for a subject by the mean of the values the subject actually has for the 30–180 minute observations and then apply the summary measure method to
for the BPRS data and summarize your conclusions.
Investigate the use of the other summary measures listed in Table
Palotie et al. (2017) investigated the “survival” or longevity of teeth restorations in Finland. The data that were extracted from the electronic patient files of the Helsinki City Public Dental
Grana et al. (2002) report the results from a nonrandomized clinical trial investigating a novel radioimmunotherapy in malignant glioma patients.The overall survival, that is, the time from the
The data are the survival times (in months) after mastectomy of women with breast cancer. The cancers are classified as having metastasized or not based on a histochemical marker. Censoring is
Fit a Cox regression model to the data on heroin addicts using the clinic as a stratifier. How do the results compare with those derived in the text?
Donati et al. (2013) studied the gambling behavior of high school students in a suburban area in Tuscany, Italy. The data set includes the gender of the 994 students and their answers to various
The data relate to a sample of girls in Warsaw, the response variable indicating whether or not the girl has begun menstruation and the exploratory variable age in years (measured to the month). Plot
The data (taken from Johnson and Albert, 2013) are for 30 students in a statistics class. The response variable y indicates whether or not the student passed (y = 1) or failed (y = 0) the statistics
The data were obtained from a study of the relationship between car size and car accident injuries. Accidents were classified according to their type, severity, and whether or not the driver was
Return to the data on do-it-yourself used in the text and use a backward search approach to assess models that allow interactions between each pair of explanatory variables. What conclusion do you
The data arise from a survey carried out in 1974/1975 in which each respondent was asked if he or she agreed or disagreed with the statement“Women should take care of running their homes and leave
arise from a prospective study of potential risk factors for coronary heart disease (CHD) (Rosenman et al., 1975).The study looked at 3154 men aged 40–50 for an average of 8 years and recorded the
The data shown in Table
are from Seeber (2005). Here, 31 patients treated for superficial bladder cancer have recorded both the number of recurrent tumors during a particular time-period after removal of the primary and the
The data shown in Table
The data were collected in a clinical trial of the use of estrogen patches in the treatment of postnatal depression. Using posttreatment depression score as the response, formulate a suitable model
The data arise from a survey of systolic blood pressure in individuals classified according to smoking status and family history of circulation and heart problems. Analyze the data using multiple
The age, percentage fat, and gender of 20 normal adults are given. Investigate multiple linear regression models with the percentage of fat as the response variable, and age and gender as explanatory
Four sets of bivariate data from Anscombe (1973) are given. Fit a simple linear regression to each data set. What do you find? Now construct regression graphics and describe what you conclude from
The data arise from a study of the quality of statements elicited from young children reported by Hutcheson et al. (1995). The variables are statement quality, child’s gender, age and maturity, how
The data were collected to investigate the determinants of pollution. For 41 cities in the United States, seven variables were recorded:SO2: SO2 content of air in micrograms per cubic meter Temp:
The data gives the average percentage memory retention measured against passing time (minutes). The measurements were taken five times during the first hour after subjects memorized a list of
The data gives marriage and divorce rates (per 1000 population per year)for 14 countries. Derive the linear regression equation of divorce rate on marriage rate and show the fitted line on a
The data gives the average vocabulary size of children at various ages.Construct the scatterplot of the data and use the scatterplot and knowledge of the data to fit a suitable model.
The data gives the final examination scores (out of 75) and corresponding exam completion times (seconds) of 134 individuals. Construct a scatterplot of the data that shows the simple linear
Reanalyze the pulse rates and heights data after taking a log transformation of pulse rate. Contrast and compare the results with those described in the text, remembering that using a log
Mortality rates per 100,000 from male suicides for a number of age groups and a number of countries are given. Construct side-by-side boxplots for the data from different age groups, and comment on
The data set contains values of seven variables for 10 states in the United States. The seven variables are 1. Population size divided by 1000 2. Average per capita income 3. Illiteracy rate (%
Shortly after metric units of length were officially introduced in Australia, each of a group of 44 students was asked to guess, to the nearest meter, the width of the lecture hall in which they were
According to Cleveland (1994), “The histogram is a widely used graphical method that is at least a century old. But maturity and ubiquity do not guarantee the efficiency of a tool. The histogram is
What is the ratio of two measurements of warmth, one of which is 25◦C and the other of which is 110◦F?
In reading about the results of an intervention study, you find that alternate subjects have been allocated to the two treatment groups. How would you feel about the study and why?
Attribute the following quotations about statistics and statisticians:a. To understand God’s thoughts we must study statistics, for these are a measure of his purpose.b. You cannot feed the hungry
You are interested in assessing whether or not laws that ban purchases of handguns by convicted felons reduce criminal violence. What type of study would you carry out, and how would you go about the
You develop a headache while working for hours at your computer. You stop, go into another room, and take two aspirins. After about 15 min, your headache has gone and you return to work. Can you
The Pepsi-Cola Company carried out research to determine whether people tended to prefer Pepsi Cola to Coca Cola. Participants were asked to taste two glasses of cola and then state which they
15.9 What is the conjoint measurement model? When should it be used? [Hint: see, e.g., Shepard, et al., 1972.]
15.8 Give an example of a metric multidimensional scaling problem. [Hint: see, e.g., Torgerson, 1958.]
15.7 In what situations is hierarchical clustering likely to be most useful?
15.6 Distinguish between clustering by subject and clustering by attribute.
15.5 Suppose all data vectors in a clustering problem are known a priori to follow multivariate Normal distributions with unknown parameters and equal covariance matrices. Devise a clustering
15.4 Is it possible to use minimum variance clustering and hierarchical clustering in the same problem? Explain your answer.
15.3 Explain how the final configuration of a multidimensional scaling problem would be altered if all dissimilarities were given as points on an interval scale and were then standardized (mean zero
15.2 In the multidimensional scaling model suppose the M dissimilarities are not known but there are N judges available to provide N sets of ordered dissimilarities. Two procedures are suggested: (a)
15.1 Show that in the multidimensional scaling model if the fitted distances ij are all within twice the preassigned distances dij, respectively, the stress is a numerical quantity confined to the
14.5 Evaluate the optional control vector for a problem in which T = 20, p = 3, q = 2, q1 = 1, w2= 2.
14.4 How would the optimal control vector setting corresponding to (14.2.17) change if a generalized natural conjugate prior were used instead of the diffuse prior used in (14.2.7)? [Hint: See
14.3 In Exercise 14.1, compute the future risk corresponding to the optimal control setting.
14.2 Determine the relationship for the optimal control vector, corresponding to (14.2.17), for the case in which the cost of control is zero.
14.1 A governmental program to assist educationally disadvantaged teenagers in improving their position in the job market provides remedial education at specially provided centers. Achievement tests
13.12 Explain how you might use the “faces” approach to classification to classify the patient’s disease in Exercise 13.11. [Hint: See Chernoff (1973).]
13.11 Suppose you were attempting to diagnose a patient’s disease on the basis of a vector of symptoms that included some variables that were continuous, and some that were discrete. Explain why
13.10 Explain the advantages of using a Bayesian classification procedure over those of a frequentist procedure.
13.9 Suppose multivariate data were arriving sequentially in time (say, velocity information about an arriving airplane, and we wish to classify the airplane as friend or foe). How should a
13.8 Use the logistic discrimination approach to develop a classification procedure for the data in Exercise 7.6 and the populations in Exercise 13.1. Classify the vectors in Exercise 13.2 by this
13.7 Suppose it were known that one of two authors wrote a particular article, but a decision as to which was the correct author was difficult. Suppose further that N1 previous articles of author
13.6 Consider the problem about radio station audiences in Example (13.4.1). Suppose instead of using the equal likelihood prior probabilities, the sample frequencies (obtained from specific
13.5 Compare the results in Exercises 13.2 and 13.4. Would you prefer one procedure over the other? Why?
13.4 Classify the vectors in Exercise 13.2 by means of the procedure developed in Exercise 13.3.
13.3 Use the Bayesian approach of Section 13.3.2 with equal misclassification costs and equal prior probabilities and apply it to the data of Exercise 7.6. Develop the predictive probability density

Showing 1 - 100 of 826