Topic: regression output paper

Order Description my part is talking about the approach i used(in word file),why is good in terms of limiting bias? and what should i improve given the data i have, also please interpretation the coefficient 'exposure, danger, interaction between dream and sleep' my partner is talking about constitutional effect that affect sleep mine is talking about environmental effect (maybe can test overall significant for those environmental variable “predation” “exposure” “danger”, ) feel free to do more researched about this topic (on environmental effect) that i used in my regression Also, I have changed some dummies on my own, according to the dataset. Like the index for danger over three is 1 lower is 0, and so on for the other variable. For dream variable, if the dream sleeping time is greater than average is 1, and lower is 0 And there is an interaction variable between dream and body weight, plz emphasize in the paper why we need that.(finding some evidence between body weight and dream sleep) Plz take a look at the STATA do file at the end ( I ALREADY HIGHLIGHT MY PART) regression Overview As environment and species protections have been a crucial focus nowadays, the analysts figured that it would be an interesting research that worths digging into on different factors that may cause fluctuating hours of mammal’s sleep. Using existing data and assumed variables, our models suggest that factors such as constitutional and ecological variables should be taken into consideration on explaining the mammal’s total hours of sleep. Such dataset provides useful comparative results and further support to our model on explaining the relationship between mammal’s sleep and all the explanatory variables. Goals ? Construct different regression models to testify the statistical significance of explanatory factors related to the mammal’s total sleeping time (dependent variable) a) use simple regression model and multiple regression model b) generate dummy variables using the existing data c) check if there are interactions between each factor (variable) d) create new interaction variables using dummy variables if existed e) make hypotheses f) use T-test and F-test to find the test statistics and compare them with critical value Approach :Previous MLR with ecological dummy variables and interaction between two variables (sophisticated) Sources ? “Paths from sleep exposure index. Exposure to predators should strongly influence sleep architecture. A sleeping mammal is essentially unconscious, and the stimulus strength necessary to awaken an animal (the arousal threshold; Tobler 2005) is much larger than that needed to alert an awake animal” (exposure and danger)(5) ? “... we see greater sleep intensity, meaning greater brain activity during sleep; dreaming is definitely increased and likely more vivid," says neurologist Mark Mahowald of the University of Minnesota and director of the Minnesota Regional Sleep Disorders Center in Minneapolis” (6) ? “REM percentage correlated significantly with body weight... total sleep time and body weight were the only factors that remained significantly correlated with total minutes of REM sleep,”(dream and bodyweight interaction)(7) STATA do file scatter total_sleep body_weight gen ln_body_weight = ln(body_weight) regress total_sleep body_weight scatter total_sleep ln_body_weight, mlab(animal) regress total_sleep ln_body_weight if animal != "Asian elephant" & animal != "African elephant" scatter total_sleep brain_weight gen ln_brain_weight =ln(brain_weight) scatter total_sleep life_span gen ln_life = ln(life_span) scatter total_sleep gestation regress total_sleep ln_body_weight ln_brain_weight ln_life gestation if animal != "Asian elephant"& animal!="African elephant" gen inter_body_dream = body_weight * dream regress total_sleep ln_body_weight ln_brain_weight ln_life gestation predation exposure danger dream inter_body_dream if animal != "Asian elephant"& animal!="African elephant"