This study has significantly contributed to Indonesian Islamic banking based on which the Islamic banking manager should recognize that the intermediation level, fee-based service activity and efficiency are crucially important in establishing competition and maintaining sustainable Islamic banking. Hardianto, D. Emerald Group Publishing Limited. Report bugs here. Please share your general feedback. You can join in the discussion by joining the community or logging in here.
You can also find out more about Emerald Engage. Visit emeraldpublishing. Answers to the most commonly asked questions here. To read the full version of this content please select one of the options below:. Access and purchase options You may be able to access this content by logging in via your Emerald profile. Rent this content from DeepDyve. Rent from DeepDyve. Although causality cannot be determined in this study, these findings suggest that multiple metabolites are involved in the etiology of cardiometabolic diseases and also influenced by the overall dietary pattern.
To our knowledge, our study is the largest study to date examining the association between the Mediterranean diet and metabolite concentrations via a metabolomics approach. With its large sample size, internal cross-validation analysis confirmed the internal validity of the metabolite score. Limitations of our work included the cross-sectional design of the Fenland Study, limiting the interpretation of the causal pathways.
For instance, metabolite concentrations could vary as a consequence of the causal effect of the diet on insulin resistance. Although we adjusted for major potential confounders, residual confounding could be present and adjustment for a metabolite in assessing mediation could cause further residual confounding Nonetheless, it is still meaningful that a mechanism of causation, confounding, or both could indeed involve specific small molecules measured by metabolomics.
Measurement errors in dietary measures, the metabolomics assays, and covariates could exist and cause both false-positive and false-negative findings for the associations of the MDS with metabolites. Another limitation is that our targeted metabolomics assay did not include some molecules relevant to the Mediterranean diet, such as dietary flavonoids.
We focused on relative differences in metabolite concentrations by adherence to the Mediterranean diet, although absolute concentrations would potentially be of interest. The assay platform we used also could not differentiate isobaric lipids with identical molecular weights. Acyl-chains of phosphatidylcholine , for example, could be both a pair of and palmitic acid and DHA and a pair of and oleic acid and EPA.
Degrees of both chain-length and -saturation influence roles of fat-esterified molecules. Therefore, roles of individual phospholipids without information on acyl-chains cannot be interpreted in this study and should be investigated in future work, for example, by using NMR spectroscopy or alternative MS platforms 45 , Collinearity between metabolites would be concerning but our cross-validation approach confirmed the robustness of the results derived from a combination of metabolites.
Because the Mediterranean diet is likely to be fit to dietary habits uniquely in different populations in Mediterranean and non-Mediterranean countries 2 , 47 , future investigation in other populations is warranted. For example, adherence to the Mediterranean diet observed in our study was not necessarily the same as that in Mediterranean countries, regarding mean consumption of olive oil, nuts, and yogurt.
Previous assessments of Mediterranean diet adherence across different regions have found that Northern European countries including the United Kingdom had lower Mediterranean diet adherence than the Mediterranean countries, but that adherence was also higher in the United Kingdom in the — period than in the — period We also did not differentiate between certain details of Mediterranean diet adherence, for example, on red wine compared with alcohol consumption, because our dietary assessment did not differentiate between red wine and white wine.
The difference could influence a combination of metabolites to predict adherence to the Mediterranean diet. Alongside the possibility that different components of the Mediterranean diet could contribute differentially between populations, the lack of replication of our findings in an independent cohort is one of the limitations of our study. Confirmatory work would ideally be conducted in both the United Kingdom and other countries.
The current findings provide evidence that different subclasses of small metabolites including carnitines, amines, and phospholipids were associated with adherence to the Mediterranean diet, and should prompt the discovery of potential biomarkers or metabolic profiles associated with this dietary pattern. In addition, the results also showed that the associations of the Mediterranean diet with measures of insulin resistance and major lipid profiles were partly explained by metabolites related to the Mediterranean diet, suggesting their involvement in pathways linking diet to disease risk.
Overall, these findings advance greater understanding of metabolites as potential dietary biomarkers and help with knowledge on pathways involved in diet—disease etiology. J Nutr. Published online Oct Author information Article notes Copyright and License information Disclaimer.
Address correspondence to FI e-mail: ku. Address correspondence to NGF e-mail: ku. This article has been cited by other articles in PMC. Objectives We aimed to investigate how the Mediterranean diet could influence circulating metabolites and how the metabolites could mediate the associations of the diet with cardiometabolic risk factors. Methods Among 10, participants Conclusions Multiple metabolites relate to the Mediterranean diet in a healthy general British population and highlight the potential to identify a set of biomarkers for an overall diet.
Keywords: Mediterranean diet, molecular epidemiology, nutritional epidemiology, biomarkers, acylcarnitines, amines, sphingolipids, phospholipids, metabolomics, dietary pattern. Introduction The Mediterranean diet is a healthy dietary pattern associated with lower risk of cardiometabolic and other noncommunicable diseases 1—3. Methods Study population The Fenland Study is a general population cohort which recruited 12, participants via general practices from 3 centers Cambridge, Ely, Wisbech in Cambridgeshire, United Kingdom 12 , Dietary assessment Dietary assessment was by a item semiquantitative FFQ which assessed dietary intake over the past year.
Targeted metabolomics Participants were instructed to fast for 10 h before their appointment time for the collection of fasting blood samples upon arrival, from which plasma samples were divided into aliquots. Assessment of cardiovascular disease risk factors and other covariates At the clinic visit, participants completed a health and lifestyle questionnaire about socioeconomic status, medication use, family history of diabetes, and smoking behavior. Statistical analyses All analyses were performed using Stata version Results Baseline characteristics Baseline characteristics of the Fenland participants are reported in Table 1 , by thirds of adherence to the MDS.
Adherence to the Mediterranean diet Baseline characteristics Tertile 1 3. Open in a separate window. Adherence to the Mediterranean diet among 10, participants recruited at baseline in — in the Fenland Study, using the dietary score derived from the Mediterranean dietary pyramid see the Methods for details; possible range: 0— TABLE 3 Contribution of groups of metabolites to the association between the Mediterranean diet and cardiovascular disease risk factors: the Fenland Study, — 1.
Regression models were fitted with adjustment for age, sex, test site, education level, income, occupation, medication use, family history of diabetes, objectively measured physical activity, smoking, BMI, and waist circumference. Discussion In a population-based cohort of adults without diabetes in the United Kingdom, 66 metabolites across different subclasses were found to be associated with adherence to the Mediterranean diet.
Interpretation of findings and implications Although many of the targeted metabolites are known to be synthesized de novo, we found that the combination of 66 metabolites had a moderate correlation of 0. Strengths and limitations To our knowledge, our study is the largest study to date examining the association between the Mediterranean diet and metabolite concentrations via a metabolomics approach.
Conclusions The current findings provide evidence that different subclasses of small metabolites including carnitines, amines, and phospholipids were associated with adherence to the Mediterranean diet, and should prompt the discovery of potential biomarkers or metabolic profiles associated with this dietary pattern. Author disclosures: The authors report no conflicts of interest. NGF and FI contributed equally to this work. References 1. Primary prevention of cardiovascular disease with a Mediterranean diet.
N Engl J Med. Prospective association of the Mediterranean diet with cardiovascular disease incidence and mortality and its population impact in a non-Mediterranean population: the EPIC-Norfolk study. BMC Med. Mediterranean diet and health status: an updated meta-analysis and a proposal for a literature-based adherence score.
Public Health Nutr. Mediterranean diet pyramid today. Science and cultural updates. Glycemic index, glycemic load, and chronic disease risk—a meta-analysis of observational studies. Am J Clin Nutr. Association of dietary, circulating, and supplement fatty acids with coronary risk: a systematic review and meta-analysis. Ann Intern Med. Nutr Metab Cardiovasc Dis. Semin Thromb Hemost. Metabolomics: an emerging post-genomic tool for nutrition.
Br J Nutr. Towards metabolic biomarkers of insulin resistance and type 2 diabetes: progress from the metabolome. Lancet Diabetes Endocrinol. Cardiovascular metabolomics. Circ Res. Genetic predisposition to an impaired metabolism of the branched-chain amino acids and risk of type 2 diabetes: a Mendelian randomisation analysis.
PLoS Med. Wishart DS. Metabolomics: applications to food science and nutrition research. Trends Food Sci Technol. Nutritional metabolomics: progress in addressing complexity in diet and health. Annu Rev Nutr. Variation of serum metabolites related to habitual diet: a targeted metabolomic approach in EPIC-Potsdam.
Eur J Clin Nutr. An overview of the role of metabolomics in the identification of dietary biomarkers. Curr Nutr Rep. Bingham SA. Biomarkers in nutritional epidemiology. Biomarkers in nutritional epidemiology: applications, needs and new horizons. Hum Genet. The association between a biomarker score for fruit and vegetable intake and incident type 2 diabetes: the EPIC-Norfolk study.
J Proteome Res. Plasma acylcarnitines and risk of cardiovascular disease: effect of Mediterranean diet interventions. Mol Nutr Food Res. Plasma lipidomic profiles and cardiovascular events in a randomized intervention trial with the Mediterranean diet. Fenland Study. Comparison of dietary assessment methods in nutritional epidemiology: weighed records v. Epidemiological assessment of diet: a comparison of a 7-day diary with a food frequency questionnaire using urinary markers of nitrogen, potassium and sodium.
Int J Epidemiol. A new tool for converting food frequency questionnaire data into nutrient and food group values: FETA research methods and availability. BMJ Open. Biocrates Life Sciences AG. AbsoluteIDQ p Kit. A genome-wide perspective of genetic variation in human metabolism. Nat Genet. Interlaboratory reproducibility of a targeted metabolomics platform for analysis of human serum and plasma. Anal Chem. Association between birth weight and visceral fat in adults.
Reliability and validity of the combined heart rate and movement sensor Actiheart. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem. HOMA-estimated insulin resistance is an independent predictor of cardiovascular disease in type 2 diabetic subjects: prospective data from the Verona Diabetes Complications Study.
Diabetes Care. Multiple imputation using chained equations: issues and guidance for practice. Stat Med. Benjamini Y, Yekutieli D. The control of the false discovery rate in multiple testing under dependency. Ann Statist. Design and analysis of metabolomics studies in epidemiologic research: a primer on -omic technologies. Am J Epidemiol. Multivariable prognostic models: issues in developing models, evaluating assumptions and adequacy, and measuring and reducing errors.
Contribution of modifiable risk factors to social inequalities in type 2 diabetes: prospective Whitehall II cohort study. Mediation analysis in epidemiology: methods, interpretation and bias. Acylcarnitines: reflecting or inflicting insulin resistance? Baseline metabolomic profiles predict cardiovascular events in patients at risk for coronary artery disease.
Am Heart J. Metabolomics signature improves the prediction of cardiovascular events in elderly subjects. Li Z, Vance DE. Phosphatidylcholine and choline homeostasis. J Lipid Res. Quantitative serum nuclear magnetic resonance metabolomics in cardiovascular epidemiology and genetics.
Circ Cardiovasc Genet.
Forums FAQ. Search in titles only. Posts Latest Activity. Page of 1. Filtered by:. Winzorised to ddress problems caused by small denominators and to control for the effect of potential outliers. Tags: None. Attaullah Shah.
Regards Attaullah Shah, PhD. Comment Post Cancel. Nick Cox. Steve Samuels. Hampel, Also, winsorizing and trimming can be bettered by other methods which adapt to likely outliers, and which do not require much of an advance guess about how many there are.
If you winsorize a variable that is destined to be the response in a regression, you probably be altering the wrong observations. You should be reducing the influence of very large residuals, not the original values. For regression, the robust regression package mmregress by Verardi and Croux is superior findit.
Once installed, you can type the following and get output similar to that above by typing just one command. We see the data for the three potential outliers we identified, namely Florida, Mississippi and Washington D. We will return to this issue later. We use the show 5 high options on the hilo command to show just the 5 largest observations the high option can be abbreviated as h. We see that DC has the largest leverage.
Here k is the number of predictors and n is the number of observations. In our example, we can do the following. As we have seen, DC is an observation that both has a large residual and large leverage. Such points are potentially the most influential.
We can make a plot that shows the leverage by the residual squared and look for observations that are jointly high on both of these measures. We can do this using the lvr2plot command. Using residual squared instead of residual itself, the graph is restricted to the first quadrant and the relative positions of data points are preserved.
This is a quick way of checking potential influential observations and outliers at the same time. Both types of points are of great concern for us. The two reference lines are the means for leverage, horizontal, and for the normalized residual squared, vertical. The points that immediately catch our attention is DC with the largest leverage and MS with the largest residual squared. These measures both combine information on the residual and leverage.
We can list any observation above the cut-off point by doing the following. DFITS can be either positive or negative, with numbers close to zero corresponding to the points with small or zero influence. As we see, dfit also indicates that DC is, by far, the most influential observation. The above measures are general measures of influence. You can also consider more specific measures of influence that assess how each coefficient is changed by deleting the observation.
We can restrict our attention to only those predictors that we are most concerned with to see how well behaved those predictors are. The names for the new variables created are chosen by Stata automatically and begin with the letters DF. The value for DFsingle for Alaska is. Since the inclusion of an observation could either contribute to an increase or decrease in a regression coefficient, DFBETAs can be either positive or negative. We add a line at.
We see the largest value is about 3. We can repeat this graph with the mlabel option in the graph command to label the points. With the graph above we can identify which DFBeta is a problem, and with the graph below we can associate that observation with the state that it originates from. The following table summarizes the general rules of thumb we use for these measures to identify observations worthy of further investigation where k is the number of predictors and n is the number of observations.
We have used the predict command to create a number of variables associated with regression analysis and regression diagnostics. The help regress command not only gives help on the regress command, but also lists all of the statistics that can be generated via the predict command. Below we show a snippet of the Stata help file illustrating the various statistics that can be computed via the predict command.
We have explored a number of the statistics that we can get after the regress command. There are also several graphs that can be used to search for unusual and influential observations. The avplot command graphs an added-variable plot.
It is also called a partial-regression plot and is very useful in identifying influential points. For example, in the avplot for single shown below, the graph shows crime by single after both crime and single have been adjusted for all other predictors in the model. The line plotted has the same slope as the coefficient for single. This plot shows how the observation for DC influences the coefficient. You can see how the regression line is tugged upwards trying to fit through the extreme value of DC.
Alaska and West Virginia may also exert substantial leverage on the coefficient of single. Stata also has the avplots command that creates an added variable plot for all of the variables, which can be very useful when you have many variables. It does produce small graphs, but these graphs can quickly reveal whether you have problematic observations based on the added variable plots.
DC has appeared as an outlier as well as an influential point in every analysis. Since DC is really not a state, we can use this to justify omitting it from the analysis saying that we really wish to just analyze states. As we expect, deleting DC made a large change in the coefficient for single.
The coefficient for single dropped from After having deleted DC, we would repeat the process we have illustrated in this section to search for any other outlying and influential observations. Finally, we showed that the avplot command can be used to searching for outliers among existing variables in your model, but we should note that the avplot command not only works for the variables in the model, it also works for variables that are not in the model, which is why it is called added-variable plot.
We can do an avplot on variable pctwhite. It is the coefficient for pctwhite if it were put in the model. We can check that by doing a regression as below. In this section, we explored a number of methods of identifying outliers and influential points.
In a typical analysis, you would probably use only some of these methods. In our example, we found that DC was a point of major concern. We performed a regression with it and without it and the regression equations were very different. We can justify removing it from our analysis by reasoning that our model is to predict crime rate for states, not for metropolitan areas. Many researchers believe that multiple regression requires normality. This is not the case. Normality of residuals is only required for valid hypothesis testing, that is, the normality assumption assures that the p-values for the t-tests and F-test will be valid.
Normality is not required in order to obtain unbiased estimates of the regression coefficients. OLS regression merely requires that the residuals errors be identically and independently distributed. Furthermore, there is no assumption or requirement that the predictor variables be normally distributed. If this were the case than we would not be able to use dummy coded variables in our models. After we run a regression analysis, we can use the predict command to create residuals and then use commands such as kdensity , qnorm and pnorm to check the normality of the residuals.
Below we use the kdensity command to produce a kernel density plot with the normal option requesting that a normal density be overlaid on the plot. It can be thought of as a histogram with narrow bins and moving average. The pnorm command graphs a standardized normal probability P-P plot while qnorm plots the quantiles of a variable against the quantiles of a normal distribution.
As you see below, the results from pnorm show no indications of non-normality, while the qnorm command shows a slight deviation from normal at the upper tail, as can be seen in the kdensity above. Nevertheless, this seems to be a minor and trivial deviation from normality. We can accept that the residuals are close to a normal distribution.
There are also numerical tests for testing normality. One of the tests is the test written by Lawrence C. Hamilton, Dept. You can get this program from Stata by typing search iqr see How can I used the search command to search for programs and get additional help? Severe outliers consist of those points that are either 3 inter-quartile-ranges below the first quartile or 3 inter-quartile-ranges above the third quartile.
Mild outliers are common in samples of any size. The residuals have an approximately normal distribution. Another test available is the swilk test which performs the Shapiro-Wilk W test for normality. The p-value is based on the assumption that the distribution is normal. In our example, it is very large.
One of the main assumptions for the ordinary least squares regression is the homogeneity of variance of the residuals. If the model is well-fitted, there should be no pattern to the residuals plotted against the fitted values. A commonly used graphical method is to plot the residuals versus fitted predicted values.
We do this by issuing the rvfplot command. We see that the pattern of the data points is getting a little narrower towards the right end, which is an indication of heteroscedasticity. Both test the null hypothesis that the variance of the residuals is homogenous. Therefore, if the p-value is very small, we would have to reject the hypothesis and accept the alternative hypothesis that the variance is not homogenous.
So in this case, the evidence is against the null hypothesis that the variance is homogeneous. These tests are very sensitive to model assumptions, such as the assumption of normality. Therefore it is a common practice to combine the tests with diagnostic plots to make a judgment on the severity of the heteroscedasticity and to decide if any correction is needed for heteroscedasticity.
In our case, the plot above does not show too strong an evidence. So we are not going to get into details on how to correct for heteroscedasticity even though there are methods available. When there is a perfect linear relationship among the predictors, the estimates for a regression model cannot be uniquely computed.
The term collinearity implies that two variables are near perfect linear combinations of one another. When more than two variables are involved it is often called multicollinearity, although the two terms are often used interchangeably. The primary concern is that as the degree of multicollinearity increases, the regression model estimates of the coefficients become unstable and the standard errors for the coefficients can get wildly inflated. In this section, we will explore some Stata commands that help to detect multicollinearity.
We can use the vif command after the regression to check for multicollinearity. As a rule of thumb, a variable whose VIF values are greater than 10 may merit further investigation. A tolerance value lower than 0. It means that the variable could be considered as a linear combination of other independent variables. All of these variables measure education of the parents and the very high VIF values indicate that these variables are possibly redundant.
In this example, multicollinearity arises because we have put in too many variables that measure the same thing, parent education. Note that the VIF values in the analysis below appear much better. This is because the high degree of collinearity caused the standard errors to be inflated. The collin command displays several different measures of collinearity.
For example, we can test for collinearity among the variables we used in the two examples above. Note that the collin command does not need to be run in connection with a regress command, unlike the vif command which follows a regress command. Also note that only predictor independent variables are used with the collin command. You can download collin from within Stata by typing search collin see How can I used the search command to search for programs and get additional help?
The condition number is a commonly used index of the global instability of the regression coefficients — a large condition number, 10 or more, is an indication of instability. When we do linear regression, we assume that the relationship between the response variable and the predictors is linear. This is the assumption of linearity.
If this assumption is violated, the linear regression will try to fit a straight line to data that does not follow a straight line. Checking the linear assumption in the case of simple regression is straightforward, since we only have one predictor. All we have to do is a scatter plot between the response variable and the predictor to see if nonlinearity is present, such as a curved band or a big wave-shaped curve.
For example, recall we did a simple linear regression in Chapter 1 using dataset elemapi2. Below we use the scatter command to show a scatterplot predicting api00 from enroll and use lfit to show a linear fit, and then lowess to show a lowess smoother predicting api00 from enroll. We clearly see some degree of nonlinearity. Checking the linearity assumption is not so straightforward in the case of multiple regression. We will try to illustrate some of the techniques that you can use.
The most straightforward thing to do is to plot the standardized residuals against each of the predictor variables in the regression model. If there is a clear nonlinear pattern, there is a problem of nonlinearity. Otherwise, we should see for each of the plots just a random scatter of points. The two residual versus predictor variable plots above do not indicate strongly a clear departure from linearity.
Winsorization is a way to minimize the influence of outliers in your data by either: Assigning the outlier a lower weight,. WINSOR: Stata module to. I would like to winsorize outliers SPSS instead of trimming them. I have 12 independent variables including the foreign exchange measures. number of outliers, which may be problematic in the large panel data studies often conducted in The use of statistical packages (Stata, SAS, etc.).