ubuntu 12.04 boot with bluetooth turned off

as I do not use a bluetooth enabled device like mouse or keyboard regularly, I did not want my bluetooth on by default. It is a simple thing to turn it off, but to do every time I log in is not worth it.

Here is what worked for me:(from http://blogs.bu.edu/mhirsch/2012/07/disabling-bluetooth-on-startup-ubuntu-12-04/)

Open the following file
sudo gedit /etc/rc.local

Before the exit 0,
add

rfkill block bluetooth

I usually add some comments if I am changing the system setting this way so I know what I have been meddling with.

dualboot Windows7 and Ubuntu

The next time I get a machine without admin rights, the first I know I should do is to install Linux as a dual boot. If you are going to use only the browser, Office and may be the media player, and are fussy about dirtying your hands at the terminal, then stick to the Windows install without admin rights. Otherwise, go for a dual boot solution with any other OS. I preferred Ubuntu as a beginner for the many reasons stated all over the web.

As of Jan 2013, Quetzal Quantal was the latest release, but since Precise Pangolin has LTS (long term support) went for PP (version 12.04). I think it is more stable and probably easier to find drivers and software.

If you decide to do create a USB bootable (for me that seemed the easiest option), check http://www.ubuntu.com/download/help/create-a-usb-stick-on-windows
download the Universal USB Installer. From then it is a breeze.

Once downloaded, start the installer, and follow the steps as in the instructions page. In the select OS tab, choose whichever desktop version you are interested in. If you have not already got a iso of the distro, check “download the iso” box, select the USB drive letter (double check this part to ensure that it is USB) and hit Create.

I needed like 30 min.

I did not do a back up of my whole Windows drive, thought it is always recommended. I think if you are trying to fiddle around right at the beginning it might be OK, but no guarantees.

Restart the system and hit one of the Fn. keys (in my case it was f12, but I think it could be f2, f5, or f8, depending on the model) to enter the BIOS. There go to Boot set-up and choose to boot from USB. Deselect boot from hard disc, to be certain. Apply the changes and the system boots into Ubuntu giving two options: to try Ubuntu or install Ubuntu.

Select try Ubuntu. For the installation, an internet connection is required. In my case, the system automatically recognised the driver needed and asks if you need to install it. Install the driver only for the internet connection, as the others that may be offered are not needed right away (like the one for Graphics accelerator which I needed to install). Now the list of available networks are shown. If using WiFi does not work at this stage, use a network cable to keep it simple.

Now the OS can be installed in the machine. Click install Ubuntu, and follow the instructions. Gparted is used to create a separate partition to install Linux. I left it at the default value. Was scared of damaging my Windows files, as I hadn’t the patience to back-up 😉

hmmm. I don’t remember the next steps exactly, but installation should proceed fine. Once done, install driver (preferably stable releases) that you are offered. Click “super” — the Windows key, and type “updates”. Select software updates from the icons that show up below, and update which ever is required. From there, you are ready to go (and learn Linux 🙂 )

saving regression results

1. use #capture.output# to save results of summaries or of models as in write.table(capture.output(summary(a model object)), ‘temp.txt’, quote=FALSE, sep = ‘\n’)

2. using the results in R-studio .Rnw files, use package TEXREG. If there are underscores, then use “\catcode`\_=12” with a T1-fontenc, to not have to explicitly escape the underscore characters everytime

testing relationships

Some variable, or set of variables, y, are predicted to have a particular relationship with
some predictor variable (or variables) denoted in x.

In the simplest case when both x and y are continuous variables, the analysis is called a regression analysis, if x has more than one predictor variable then it is called a multiple regression, and if y is binary it is a logistic regression.

However, if the predictor variable is categorical the model is called an analysis of variance with many variants depending upon the number and relationship of categorical predictor variables in x. Finally, if predictor variables consist of categorical and continuous variables then it is called an analysis of covariance

Linear probability model (LPM) is one where the probability of an event occuring is tested. the dependent variable is binar.

Linear regression is suitable for measurement var, but not suitable for nominal variables like ratios and proportions either. Cobning measurement vars with nominal vars in a regression test, would produce erroneous results. In the case of nominal vars, logistic regression is more appropriate