Help Taking A Derivative... I Know How To Just Need the Function =|


by GreenPrint
Tags: derivative, function
GreenPrint
GreenPrint is offline
#1
Sep26-10, 08:48 PM
P: 1,184
Ok I'm doing this lab and have the following data
the only problem is that the lab requires me to take the derivative of the function that the data is suppose to represent, so sense my data is not exact, i.e. the function that it's suppose to model might have f(3)=2 but sense i did a lab and collected data on for various points for the function my data might be f(3)=2.789 etc. So what I believe I need to do is using my data come up with a approximate function that my data is suppose to represent using regression analysis... can someone please help me with this below I've included my data thanks

x...........................y
0.01.......................0.162140251
0.013333333............0.121396055
0.02.......................0.112076212
0.04.......................0.698080279
0...........................infinity

i.e. my data (.01, 0.081070126) etc.

THANKS!!!!
Phys.Org News Partner Science news on Phys.org
NASA's space station Robonaut finally getting legs
Free the seed: OSSI nurtures growing plants without patent barriers
Going nuts? Turkey looks to pistachios to heat new eco-city
rock.freak667
rock.freak667 is offline
#2
Sep26-10, 08:53 PM
HW Helper
P: 6,213
What regression model did you use? (As in what equation did whatever software you used come up with?)
GreenPrint
GreenPrint is offline
#3
Sep26-10, 08:56 PM
P: 1,184
I haven't used a regression model I have forgeton how to do it by hand =| the sad thing I took statistics a year ago and already forgot it =(

fzero
fzero is offline
#4
Sep26-10, 08:58 PM
Sci Advisor
HW Helper
PF Gold
P: 2,606

Help Taking A Derivative... I Know How To Just Need the Function =|


Without knowing what you're supposed to be doing with the derivative, it's tough to hazard a guess as to whether or not what you're proposing is appropriate. Have you considered whether your analysis should actually use your data to approximate the derivative at the midpoints between your data points? Your data gives a coarse rate of change over the values at which you made measurements.

Note, doing regression analysis is usually meaningless unless you already have a hypothesis for what function should describe the data.
GreenPrint
GreenPrint is offline
#5
Sep26-10, 09:00 PM
P: 1,184
Well I was doing a biology lab and the data we collected was crap... it asked me to find the rate of reaction... I think it's suppose to be nearly a x^.5 function

let me go get more information on this
rock.freak667
rock.freak667 is offline
#6
Sep26-10, 09:02 PM
HW Helper
P: 6,213
Quote Quote by GreenPrint View Post
Well I was doing a biology lab and the data we collected was crap... it asked me to find the rate of reaction... I think it's suppose to be nearly a x^.5 function

let me go get more information on this
Normally in a rate of reaction lab, well the ones I did for chem (don't know if bio differs), I never encountered a point such as x=0, y=∞.
GreenPrint
GreenPrint is offline
#7
Sep26-10, 09:19 PM
P: 1,184
Well before this was a biology question as I had no idea what the heck I was suppose to do but then I figured it out and now am stuck on the calculations part

In this lab I took a 50 mL beaker and poured in 10 ml of cold distilled water into it, I then poured in 30 ml of original enzyme solution (I believed we used cold potato juice or something), I then took a piece of paper and placed it into a beaker of this concentration, 75%, after waiting for 5 seconds i then took the 2.1 cm filter paper disc from the 75% concentration and placed it onto a piece of paper, after draining for 10 seconds I then placed it into my solution of hydrogen peroxide at the bottom of a 50 mL beaker then removed the forceps i was using to hold the paper and timed the time it took for the paper to rise from the bottom of the 50 mL beaker to the surface and got .83 seconds, how do I calculate the rate of reaction

THANKS!!!!!
by the way I know
the formula is
(1/q) ( (d[Q])/(dt) )
wikipedia is very useful
http://en.wikipedia.org/wiki/Reaction_rate


My teacher told me that the rate of reaction = 1/t which I think is just bs because that would be the frequency of the paper as it rose...

so I did some research and from the equation above it is clear that I needed to graph my data, my original data was

enzyme concentration........tiral 1..... trail 2.... trial 3.....trial 4
100.................................1...........0.83........9.64......1 3.2
75...................................2.02.......1.81........5.93....... 23.19
50.....................................1.........1.11........10.46..... ..23.12
25......................................0.97......1.04......2.37....... ..1.35
0....................................infinity.....infinity....infinity. .....infinity

so by the looks of it my data is crap oh well I figured I wouldn't fudge the data as your not suppose to and just go on with the lab report

so it took me a while to figure it out but apparently I'm suppose to graph 1/t on the y access and 1/concentration on the x axis and take the average first derivative of the function from the data

change of y/change of x = (1/t)/(1/concentration) = concentration/t which agrees with the definition of rate of reaction (1/q) ( (d[Q])/(dt) )

so I took the average of the four trails and took the inverse and used that for my time for that concentration, i.e. 1/((1+0. 83+9.64+13.2)/4) = 0.162140251, the time I'm going to use for 100% concentration

so then I was like well whats q sense it's catalase and is technically speaking not part of the reaction, then i figured out that it was 1 =|

H2O2 -----catalase----> 2H2O + O2

were catalase is suppose to be above the arrow

so as you can see I just need the function
GreenPrint
GreenPrint is offline
#8
Sep26-10, 09:27 PM
P: 1,184
Ok sorry about that I got two things mixed up and some of my english was bad and I fixed that I also fixed the A1 A2 in the formula sorry about that I just copied the data form excel and it copied the equation in cell format

but now everything should be understandable let me know if you don't understand
GreenPrint
GreenPrint is offline
#9
Sep27-10, 05:12 AM
P: 1,184
let me know if you had no idea at all what i was doing
HallsofIvy
HallsofIvy is offline
#10
Sep27-10, 07:39 AM
Math
Emeritus
Sci Advisor
Thanks
PF Gold
P: 38,882
Well, the problem is that you are telling us that you [b]measured[b] a value of infinity! I don't believe it!
GreenPrint
GreenPrint is offline
#11
Sep27-10, 12:53 PM
P: 1,184
Ok so I observed did not rise then so I guess I should consider this to be zero than, so my question is how do I find the rate of reaction

I'm also woundering if i should have considered the concentrations as
100% as 1
75% as 3/4
50% as 1/2
25% as 1/4
0% as 0 instead?

here's the original lab thing if you want it
http://webcache.googleusercontent.co...ient=firefox-a
it's a google chace version so you don't have to download it i was doing the effect of enzyme concentration on reaction rate

i can tell from the formula that raising the concnetraion increases activity just from insepction of the formula I'm just having problems coming up with the function in order to take the derivitive of it based of the data I collected

THANKS!@!!!!!1
GreenPrint
GreenPrint is offline
#12
Sep27-10, 06:39 PM
P: 1,184
can someone please help me?
Mark44
Mark44 is offline
#13
Sep27-10, 07:23 PM
Mentor
P: 21,012
Quote Quote by GreenPrint
x...........................y
0.01.......................0.162140251
0.013333333............0.121396055
0.02.......................0.112076212
0.04.......................0.698080279
0...........................infinity

i.e. my data (.01, 0.081070126) etc.
What is what you're calling "my data"? I see an x value of .01, but the y value isn't .08107...etc.

Since you are not at all clear on what you're supposed to do, you might just calculate the average rate of change between each pair of successive x-values. That would be [itex]\Delta y/\Delta x[/itex] for each pair of successive x- and y-values. Ignoring that .013333 for the moment, if you used .01 and .02, you could get approximations to f'(.015) and f'(.03).

If you are actually supposed to use regression, you're going to have to find out what sort of regression - linear, quadratic, cubic, log, whatever - you need to use.

BTW, your data does not look like it fits a square root function, as you asserted earlier in this thread.


Register to reply

Related Discussions
Taking the first derivative of a function with improper integral Calculus 10
Taking the derivative Calculus & Beyond Homework 1
Taking a derivative? Calculus & Beyond Homework 2
taking the derivative Calculus & Beyond Homework 2
taking derivative Calculus & Beyond Homework 2